Opencode-native has the same process-leak shape codex-native did (fixed in
#3925): each session runs a runner-owned `opencode serve` subprocess tracked
in _AUTO_OPENCODE_SERVERS plus the opencode TUI pane. Only DELETE /v1/sessions
cancelled the forwarder (whose finally closes the server); the other ways the
TUI pane goes away left the server orphaned for the runner's lifetime:
- the idle pane reaper closed the tmux pane but never touched
_AUTO_OPENCODE_SERVERS,
- an unexpected TUI exit (crash / OOM / host recycle) evicted the pane
without cancelling the forwarder, and
- a graceful host/runner stop tore the runner down without a per-session
DELETE, so _stop_pm never closed the servers.
Mirror the codex fix: add teardown_opencode_native_server (cancel the
forwarder, close any leftover registered server; no-op when none is
registered) and teardown_all_opencode_native_servers (shutdown sweep). Wire
them into the idle-reaper reap, the terminal-exit publisher, and _stop_pm
alongside the codex calls.
No boot-time reconcile: opencode has no crash-safe process registry and
`opencode serve` is a plain Popen (not start_new_session=True), so it shares
the runner's process group and dies with a hard runner death — the
graceful-stop + reaper + exit paths cover the observed leak.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
omnigent-telemetry#15 introduces a CloudFront default config
(omnigent_version: "default") served for any version that lacks an
explicit config file. Without this change, the version check on line 190
always rejects the default payload and silently disables telemetry.
Accept "default" as an equivalent of the current VERSION so the default
config is honoured.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
The historical-replay redaction (_redact_inline_base64) only matched
whole-string "data:*;base64,..." URIs — the resolver form under
image_url / file_data. But Claude Code's Read tool returns an image file
as an Anthropic content block {"type":"image","source":{"type":"base64",
"data":"..."}} — raw base64 with no data: prefix — carried in a
function_call_output. That shape slipped past redaction, so if it reached
the "Conversation so far:" text prefix json.dumps flattened the full
base64 into prompt text (the same class of overrun that wedges resume on
the native path).
Extend _redact_inline_base64 to also rewrite image/document base64
"source" blocks to a compact "[image/attachment: <media>, <N> base64
chars]" placeholder. Verified: image and document source blocks now
redact (base64 absent), data-URI and plain-text paths unchanged.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
The per-turn idle watchdog fails a turn that emits no non-heartbeat
events for the window. Context compaction's summarizing LLM call runs
as a single long await that emits nothing until it returns, so on a
near-full context it can exceed the 240s default and trip the watchdog.
That wedges the session in a "Prompt is too long" -> compaction ->
240s-timeout loop, since every retry re-triggers the same slow compaction.
Raise the default from 240s to 600s so a healthy long compaction has
room to finish. The HARNESS_TURN_TIMEOUT_S env knob and the absolute
ceiling are unchanged.
Co-authored-by: Isaac
* fix(web): persist open shell tabs per session
Shell tabs lived only in transient component state and the
conversation-switch effect cleared them on every navigation, so opening
a shell, switching sessions, and returning lost the tab. The PTYs
themselves live on the server and are re-fetched by useTerminals — only
the tab strip was being discarded.
Persist openTerminals/selectedTerminalKey per session in
sessionWorkspaceState (mirroring the open file tabs), seed and restore
them on mount/switch, and gate the dead-tab prune effect on the
terminals list's loading state so a restored tab isn't wiped by the
transient empty list before the session's terminals load.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e): cover shell-tab persistence; skip prune on errored terminal fetch
Add an e2e_ui test that opens a real shell in one session, switches to
another via the sidebar (client-side nav), and returns — asserting the
shell tab and its live PTY are restored. This exercises the
conversation-switch effect that regressed, which a full page reload
wouldn't.
Also address review feedback: the dead-tab prune effect ran whenever the
terminals query wasn't loading, but an errored fetch also yields an empty
list — a non-authoritative one. Pruning against it would wipe restored
tabs whose PTYs we simply couldn't reach. Gate the effect on
terminalsError as well, with a component test for the errored-read case.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Claude Code >= v2.1.197 writes `status: "shell"` to its per-session status
file when a turn ends but a background shell is still alive. The status-file
poller's map didn't know that literal, so `read_session_status` returned
`None`, the poller fired no edge and stayed stuck on its last `running` (while
also suppressing the PTY watcher's `idle`). The session never reported idle
while a background shell ran, so `sessionStatus` stayed `running`,
`shouldQueueSend` returned true, and every new message queued client-side —
regressing the "don't queue while only background work runs" behavior.
Map `shell` to `idle`: the agent loop is idle, and the Stop hook separately
relabels its own `idle` to `waiting` with the shell tally, which is what keeps
the "N background tasks still running" spinner lit.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
This reverts commit 617293d3d9.
Painting a cached transcript before revalidation meant the contents
moved under the reader: the window appeared instantly, then shifted as
newer commits were gap-bridged onto it. A hydrate spinner that resolves
into a settled transcript reads better than a fast paint that jumps, so
go back to the cold-load spinner on every conversation switch.
Co-authored-by: Isaac
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* ci: mirror linked issue priority onto closing PRs
Add a workflow that copies an issue's priority label (P0-P3) onto the
PR that closes it. Only closing links (closes/fixes/resolves #n) count;
a plain "related to #n" mention is ignored. When a PR closes several
issues the highest priority wins, and stale priority labels are dropped.
Runs on PR events and re-syncs when an issue's priority label changes;
the issue-label trigger is gated to priority labels only so other label
edits don't spin up the job.
Co-authored-by: Isaac
* ci: address review feedback on priority sync
- Tolerate null GraphQL nodes (unknown PR number, data: null) instead of
crashing on AttributeError; cover the parsing with tests.
- Add a 30s urlopen timeout so a stalled connection fails fast.
- Validate PR_NUMBER is an integer with a clear message.
- Surface a warning when the issue->PR GraphQL lookup fails rather than
silently succeeding.
- Pass the resolved PR list through an env var instead of interpolating
it into the run block.
Co-authored-by: Isaac
Rename the `enhancement` label to `Feature` and `documentation` to `Docs`
across the issue-triage system. The triage agent's `type` value is applied
verbatim as an issue label, so update the validator allow-list, the agent
schema and classification rule, the feature-request template's auto-label,
and the design proposal doc to keep them coherent.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* perf(host): cache auth headers and parallelize status payloads
Two follow-on speedups for omni host status:
1. Cache _remote_headers() per base_url within a process.
Databricks SDK credential resolution (~3s) ran on every
_host_http_json call. Since tokens are valid for the lifetime
of a CLI invocation, resolving once and reusing is safe.
A threading.Lock serialises concurrent first-time resolution
for the same URL.
2. Build daemon status payloads in parallel with ThreadPoolExecutor.
With the dead-process skip from the previous commit, only live
daemons make HTTP calls. Parallelising them lets independent
servers be queried concurrently instead of sequentially.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* chore: restore uv.lock to main
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: move header cache resolution inside try/except in _host_http_json
_remote_headers() does file I/O and Databricks SDK calls that can raise
OSError. The cache-populating call was outside the try block, so such a
failure propagated unhandled. Under ThreadPoolExecutor (added in this
PR) that aborted the entire omni host status listing.
Move the resolution inside the existing try/except so auth/file errors
remain recoverable and produce a status_code=0 result per daemon,
matching the pre-change behaviour.
Also adds test_host_http_json_handles_remote_headers_oserror to pin
this contract.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* chore: fix import order (ruff)
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* ci(triage): re-triage issues when needs-info is cleared
Add a hybrid needs-info lifecycle. When the issue author comments on an
issue that still carries needs-info, needs-info-response.yml removes the
label using the omnigent-ci App token (the default GITHUB_TOKEN would not
re-trigger downstream workflows). That removal fires issue-triage.yml's
new `unlabeled` trigger, which reads the reporter's follow-up comments,
reclassifies, and assigns an owner — re-adding needs-info only if the
issue is still too vague. Issues the reporter never clarifies are closed
by the existing stale.yml.
issue-triage.yml changes:
- trigger on issues [opened, unlabeled]; the unlabeled path fires only
for needs-info on an open issue, and allows a bot actor (the App)
- feed the author's follow-up comments into the triage prompt
- remove needs-info on re-triage when the LLM no longer flags it
- suppress the duplicate-of comment on the re-triage path
- add a per-issue concurrency group
Co-authored-by: Isaac
* ci(triage): address review — idempotent label removal, dormant-App notice
- needs-info-response.yml: re-check live labels before `gh --remove-label`
so a stale event payload / race can't fail the step (gh errors on a
missing label); emit a ::notice:: when the omnigent-ci App is
unconfigured so a dormant feature is distinguishable from a broken one.
- issue-triage.yml: also suppress the `duplicate` label on the re-triage
path (not just the comment), keeping the label and its explanation
consistent; hoist `import os` to the top of the block.
Co-authored-by: Isaac
* feat(webui): capture raw SSE events and show in execution logs panel
- sseEventLog.ts: module-level ring buffer (max 500 events/session)
with subscribe/snapshot API for useSyncExternalStore
- useSseEventLog.ts: React hook that subscribes to the ring buffer
- chatStore.ts: tap tapSessionEvents to push each StreamEvent into the
ring buffer; clear on fresh stream bind (not reconnect)
- ExecutionLogsPanel.tsx: add Items/SSE toggle — SSE tab shows
timestamped raw events with expand-to-pretty-print, auto-scrolls
to bottom as events arrive
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* perf(webui): skip SSE ring buffer when debug mode is off
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* perf(webui): cache isDebugMode as module-level boolean
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(webui): return new array ref on push so useSyncExternalStore re-renders
Object.is on the same mutated array always returns true, causing React
to skip re-renders. Produce a fresh array on every push/trim so the
snapshot reference changes and the SSE list updates in real time.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(webui): support localStorage debug flag in addition to ?debug=1
Both useDebugMode and the SSE ring buffer guard now check
localStorage.getItem("debug") === "1" as a fallback, so debug mode
can be toggled once in the console without keeping ?debug=1 in every URL:
localStorage.setItem("debug", "1") // enable
localStorage.removeItem("debug") // disable
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(webui): stable snapshot ref and correct debug flag detection
- snapshotSseLog: return shared EMPTY constant instead of allocating a
new [] on every call; prevents useSyncExternalStore render-loop from
the unstable reference on sessions with no log yet
- isDebugMode: re-read window.location.search + localStorage on every
call instead of caching against popstate; React Router uses pushState/
replaceState which never fires popstate, so the cached value stayed
stale when navigating to ?debug=1 in-app
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(web): split the harness picker by support level
The landing composer's harness picker split its primary list and "More"
group by host readiness, so any configured harness led: Claude Code,
Codex, Cursor, and Pi all competed for the few primary slots, while "More"
held only harnesses that happened to need setup. Support level — what
actually distinguishes these integrations — wasn't represented at all.
Add a `fullySupported` flag to `NativeCodingAgentSpec` and set it on
Claude Code and Codex, the integrations we maintain and test end to end.
Only those lead; every other harness folds into "More" whether or not it
is configured on the host. The flag is opt-in, so the supported set is two
lines in one file rather than a marker on each of the nine others, and a
test asserts the set is exactly claude + codex so it can't drift silently.
Two behaviors are preserved: selecting a harness pins it inline via the
existing `effectiveAgentId` rule, so the active pick is never buried; and
the hide-unconfigured preference still outranks support level, dropping
harnesses that can't launch here (and the "More" trigger with them when
that empties the group).
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(web): promote previously-launched harnesses in the picker
Splitting the picker by support level left Pi and Cursor users a hover
away from their harness on every new session, even though the split is
right for a first-time user. Nothing recorded which harnesses someone
actually launches.
Add a localStorage-backed `useRecentHarnesses` (modeled on
`useRecentWorkspaces`, but not host-scoped — a preference for Pi follows
the person across machines) and record the canonical harness id on a
successful create. The picker then promotes any recorded harness into the
primary list alongside the fully supported ones, so a regular Pi user
gets one click instead of one hover, while a fresh install still leads
with Claude Code and Codex only.
Recording happens only after the create succeeds, so a harness the user
merely browsed past never earns a slot, and the hide-unconfigured
preference still outranks recency: promotion applies within what can
launch on the host, never resurrecting a harness that can't run there.
Stored ids fold through the reversed-alias map, so `native-pi` matches
the canonical `pi-native` spec.
Also fixes the two CI failures from the support-level split: the flow
test's `selectAgent` helper now drills into "More" only when the row
isn't already inline, and the harness-install e2e no longer drills for
Codex (fully supported, so it leads inline even while needing setup).
Adds tests/e2e_ui coverage for both behaviors, stubbing every harness as
configured so the split is provably driven by support level rather than
host readiness.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(webui): wire SessionRail into AppShell behind ?debug=1
SessionRail and ExecutionLogsPanel were implemented but never rendered.
Add SessionRail as a desktop-only column between the chat and workspace
panel, gated on debugMode so it only appears with ?debug=1. The column
hides automatically when a push panel (terminals or execution logs) is
open.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(webui): remove TerminalsCard from SessionRail debug rail
Terminals are already shown in WorkspacePanel. The debug rail should
only show the Execution logs card. Also removes the onExpandTerminals
prop and all terminal-related dead code from SessionRail.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(webui): fix execution logs card title overflow in debug rail
Widen the debug column from w-48 to w-56 and add truncate/min-w-0 to
the CardTitle so the text doesn't overflow into the action buttons.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(webui): add top padding to debug rail column
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(web): keep chat content clear of the TurnRail as the area narrows
PR #4085 replaced the transcript's md:pl-12 left inset with a symmetric
px-4 gutter, dropping the clearance that kept the centered chat column off
the left-edge TurnRail (the tick minimap). On a narrow conversation area
the prose crowded the ticks.
Restore the clearance as a continuous, width-driven clamp keyed on the
conversation area (@container/chat) rather than the viewport: the column
slides left with the area until its edge nears the rail, then the left
inset ramps up to hold a minimum gap and caps at 3rem so it stops moving
instead of snapping. Because it reads the area width, opening the sidebar
feeds it too.
Add a multi-turn visual-snapshot test that mounts the rail (it only renders
for >= 2 turns, so the one-turn baseline never covered it), rendered at a
narrower viewport so the inset is actually engaged in the capture.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e-ui): regenerate visual baselines
* fix(web): shrink rail gap to 24px and stop the pill leaking into snapshots
Reduce the restored TurnRail clearance cap from 3rem to 1.5rem (24px) so the
column sits closer to the ticks while still clearing them.
Park the pointer out of the transcript's top hover band before capture in both
chat snapshot tests. Playwright's virtual mouse starts at (0,0), inside the band
that reveals the "Jump to top" pill (and, on the rail test, over a tick), so a
load-timing race could flash that transient chrome into the resting-state
baseline. Moving the pointer low pins it hidden.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e-ui): regenerate visual baselines
* test(e2e-ui): hide the Jump-to-top pill from chat snapshots deterministically
The pill is transient chrome: the initial layout settle (LatestTurnSpacer +
StickToBottom pinning to the bottom) fires a scroll that reveals it for ~2s, so
whether it lands in a capture is a race — which is why a regenerated baseline
picked it up. Force it hidden via an injected style, the same way the shared
settle kills the blinking caret, so the resting-state baseline is deterministic
regardless of when the scroll settles.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e-ui): regenerate visual baselines
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
- PiExecutor._resolve_model: strip trailing [1m]-style bracket suffixes before
passing model IDs to the Databricks AI Gateway. The direct Anthropic API
accepts e.g. system.ai.claude-opus-5[1m] but the gateway endpoint does not
(returns 404).
- CodexExecutor.run_turn: when model_provider_override is set (cli-config path)
pass model=None to thread/create so the codex binary uses its own configured
model rather than forwarding an unresolvable alias (e.g. gpt-5.6) to the UC
API.
- credential_label: cli-config providers now label from the entry name
(provider_display_name) rather than the display_name field, for consistency
with other provider kinds.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* Root cause fix — omnigent/policies/builtins/_shell.py
sudo/env/command/time/exec moved out of CMD_WRAPPERS (skip-one-word) into _FLAG_WRAPPERS with their value-consuming flags. CMD_WRAPPERS is now just {"nohup"}, which genuinely takes no options. -- needs no entry — it's consumed as a valueless flag.
While verifying, I found the same hole one level down, which also affects the original GHSA-fixed wrappers: _skip_flag_wrapper_args matched value flags by whole-token equality, so bundled short options bypassed too — sudo -nu root git push, env -iu FOO git push, and (pre-existing) nice -qn 10 git push. It now scans the bundle's characters and consumes a separate value only when the value-taking option is the bundle's last character, so -n 10/-o L still consume while -n10/-oL stay attached. This mirrors orchestration.py:236-245, which already got this right for blast_radius.
Fail-safe backstop — new is_unresolved_invocation(), wired into both consumers
The wrapper tables are an enumeration, so I didn't want the next unmodelled wrapper to be another silent ALLOW. A head still starting with - now routes through each policy's existing "can't parse this" path rather than abstaining — ASK in github.py, the configured action in working_dir.py. Reachable today via nohup -- git push …. Detection is shared; the response stays per-policy, per the module's stated contract.
* 1. env -S / --split-string (the blocker). Reviewer was right: modelling -S as a value flag swallowed the command into the flag's value, leaving zero tokens — which is_unresolved_invocation([]) can't see. Fix takes the reviewer's option (b): env -S is a command interpreter like sh -c, so it's unwrapped and re-parsed on the path that already exists for bash -c / eval.
- _skip_flag_wrapper_args gained a capture_flags set and now returns (index, captured) — reusing the existing flag walk (which already handles --flag=v, -S v, -Sv, bundles like -iS v) instead of writing a second scanner.
- real_invocation_tokens stops at env when a split-string is captured; unwrap_shell_command returns it → recursion gates the inner command.
env -S 'git push <evil> main' → DENY. env -S 'npm test' → still abstains.
2. /usr/bin/sudo -u root git push — same fail-open, not flagged in either review. Wrapper lookup matched the bare word only, so a path token became the apparent command and the segment abstained → ALLOW. Wrappers now match on basename (unwrap_shell_command already did).
* fix(policies): add BSD sudo -a/--auth-type and -c/--login-class to value-flag set
These two options were missing from _FLAG_WRAPPERS["sudo"], leaving a
residual silent-ALLOW bypass: sudo -a foo git push ... left "foo" as
the apparent command head, which does not start with "-" so is_unresolved_invocation
could not catch it. Add both flags and tests for each form.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
Co-authored-by: Tomu Hirata <tomu.hirata@gmail.com>
test_scheduled_task_create_edit_modal_and_time_picker flaked ~30% of runs,
always timing out on `_pick_minute`'s `name_input.click()` with
"dialog-overlay intercepts pointer events". While the time-picker Popover is
open, the Radix Dialog owns pointer hit-testing over the modal, so a normal
actionability-gated click at the input's coordinates resolves to the overlay
and blocks the full 30s under load.
Force every dismiss click on the name input (`click(force=True)`) — the same
technique the picker's open click already uses. A forced click still
dispatches a real pointerdown on the input, which Radix registers as the
interaction-outside that closes the popover, without waiting on overlay
actionability. Covers all three dismiss sites: the retry path and final
dismiss in `_pick_minute`, plus the two post-typed-time blurs in the test body
(focusing the time input reopens the picker via onFocus).
Verified: reproduced the flake (multiple failures across batches of 5-8 runs),
then 12/12 green after the fix; the full file's 9 tests pass.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* refactor(web): decouple typography from interface geometry
Make the desktop font preference drive semantic text tokens while keeping icons, controls, and spacing fixed.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* refactor(web): migrate interface body text to text-ui
Use the settings-controlled semantic body token across shared components and application pages for consistent sizing.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* style(web): refine sidebar typography and empty states
Align sidebar hierarchy with settings-controlled tokens and make empty projects easier to scan.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* style(web): tighten sidebar density and theme polish
Unify sidebar row geometry, refine theme-specific colors and canvas treatments, and standardize compact controls.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* test(e2e-ui): align font size checks with typography tokens
Update browser assertions for the discrete desktop font token and its current bounds.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* test(ui-snapshot): update typography visual baselines
Adopt the CI-rendered snapshots for the intentional settings-driven typography changes.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* fix(web): preserve dark active sidebar hover
Keep selected row colors stable when hovering in dark mode across both sidebars.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* style(web): polish sidebar actions and overlays
Align sidebar controls, dropdowns, and tooltips with shared density, typography, and interaction tokens for a more consistent visual hierarchy.
* style(web): normalize mobile sidebar scale
Keep mobile sidebar typography and icon geometry predictable without changing the desktop presentation.
* style(web): refine responsive sidebar and chat density
Use responsive sidebar spacing and settings-driven chat typography so mobile and desktop retain clear, consistent reading rhythm.
* test(web): align CI expectations with sidebar polish
Update E2E assertions and reviewed visual baselines to reflect the intentional typography, navigation, and density changes.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
---------
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* dev/repro-agent: pin the verdict handoff to a single JSON block
The output contract only said "a single structured verdict block" without
pinning a format, so the agent rendered YAML on some runs and JSON on others,
and the shape drifted (missing facets, prose bullets instead of objects). That
makes the `verdict` field — which the caller parses to label the issue —
unreliable to extract.
Pin it: exactly one fenced ```json block as the final message, JSON only, every
key always present, and `verdict` restricted to the four lowercase literals so
it matches verbatim. `facets` becomes an array of {symptom, verdict, evidence}
objects instead of free-form bullets. README step 4 updated to match.
Co-authored-by: Isaac
* dev/repro-agent: require the JSON block be the last chunk, allow prose above
Some runs split the artifacts into separate markdown sections (a small
"Reproduction Verdict" block, then prose "Journey"/"Facets" headers) with no
single consolidated handoff, so there was no reliable last block to parse.
Clarify the contract: comprehensive prose above the block is fine, but the
```json block must be the LAST chunk of the final message (nothing after its
closing fence) and must carry the complete self-contained handoff. Explicitly
forbid splitting the artifacts across separate sections/headers. There is no
output-schema enforcement for the claude-sdk agentic loop (AgentSpec.output_type
is inert), so this is enforced by instruction plus last-```json-fence parsing on
the caller side.
Co-authored-by: Isaac
The "new session in project" pencil navigates to /?project=<name> while
the landing screen stays mounted. The project-prefill state machine only
restarted when the ?project= param changed, so re-clicking the SAME
project's pencil after editing its default settings kept the stale seeds
— the fix only showed up after clicking another project (or Home) and
back, which flipped the param away and back.
Track a signature of the config the machine last settled from and restart
the prefill when that content changes for the same project, mirroring the
project-switch reset. The saved config is already fresh in the react-query
cache; this makes the machine re-read it.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* Add deep-research example (single agent over an MCP search server)
A single-agent example that answers a question with a cited, cross-checked
report: it plans sub-queries, searches the live web and reads full pages
through an MCP search server, and verifies claims across independent sources.
It is the repo's first example that wires an MCP server via tools/mcp/*.yaml
(auto-discovered), so it also documents the MCP extension path. One agent plus
one MCP server, no sub-agents — the simplest example to copy from. Runs
zero-config against a public, keyless endpoint.
Signed-off-by: Ilya Bogin <ilya.bogin@keenable.ai>
* test: add e2e coverage for the deep-research example agent
The examples-coverage-sync drift guard (test_every_agent_has_a_dedicated_test_file)
requires every example agent to have a dedicated e2e test. The deep-research
example shipped without one, failing E2E Tests (shard 0/4).
Add a structural test via validate_agent_def_structure (infra-free: the agent's
tools come from the hosted Keenable MCP server and it runs on the claude-sdk
harness, so it can't run end-to-end in CI). Because the agent name 'deep-research'
has a hyphen (not a valid Python test-module name), the test lives in
test_deep_research_example.py and the guard is told via a 'deep-research' entry
in _ALT_COVERED, mirroring the existing 'openai-coder' handling.
Signed-off-by: Ilya Bogin <ilya.bogin@keenable.ai>
* docs: show deep research search provider options
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Ilya Bogin <ilya.bogin@keenable.ai>
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Co-authored-by: Pat Sukprasert <pattara.sk127@gmail.com>
Co-authored-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* feat(qwen,goose): record delegated fs I/O and gate it with result-phase policy
Omnigent's OSEnvironment but left two layers as documented follow-ups: the
delegated I/O was invisible in history and no content policy ran on it.
Wire both onto the existing _handle_fs_read / _handle_fs_write handlers:
- emit a paired ToolCallRequest + ToolCallComplete per op so the I/O shows in
history (the adapter renders them as observed function_call items)
- run PHASE_TOOL_RESULT content policy on the bytes; an explicit deny refuses
the op (a write is gated before it happens), failing open otherwise
Content-only: the harness policy round-trip carries no request_data, so the
payload is {"result": content}. Closes the file-I/O recording / content policy
item in docs/QWEN_FOLLOWUPS.md.
Signed-off-by: Enes Yilmaz <enesyilmaz5157@gmail.com>
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(qwen,goose): gate delegated fs at the call phase and audit stale ops
Addresses the review on the delegated-fs recording/policy work.
1. Phase semantics. A delegated write was gated by a result-phase policy eval
before the write, which is content-only and fails open, so a policy timeout
would let the write through. Gate writes (and reads) at PHASE_TOOL_CALL with
the tool name, path, and content, failing closed on an eval error or an ASK
verdict (delegated fs has no elicitation path). Reads keep the result-phase
content check that decides whether the read bytes reach the model.
2. Audit records. Stale prior-turn server fs requests were answered at turn
start, running real I/O, and then had their ToolCall events cleared before
they reached history. Drain those events into history instead of dropping
them, so the I/O they performed is recorded.
Signed-off-by: Enes Yilmaz <enesyilmaz5157@gmail.com>
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(qwen,goose): evaluate result-phase policy after a delegated write
The write handlers gated at PHASE_TOOL_CALL and then wrote, but never ran a
result-phase evaluation, so the value env.write() returned was never policy
checked and the audit record dropped it. Reads already did both phases.
Run PHASE_TOOL_RESULT after the write carrying the actual result. A denial
records BLOCKED and refuses the response; it cannot undo the write, since it
runs after the operation. The success record now carries the real result too,
matching the read path.
_fs_content_policy_denies was read-specific, so it is now
_fs_result_policy_denies and takes any result. Read behavior is unchanged.
Signed-off-by: Enes Yilmaz <enesyilmaz5157@gmail.com>
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Enes Yilmaz <enesyilmaz5157@gmail.com>
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
PR #3105 removed the `server start` subcommand in favor of
`server --background` and updated the Electron shell-out in the same
commit. The desktop app ships on its own electron-updater channel, so a
client built before v0.7.0 is a normal steady state against a v0.7.0
CLI — and it still runs `omni server start`, which now dies with
"No such command 'start'". "Start locally" is broken for those users.
Restore the subcommand as a hidden alias that routes to the same helper
as the flag, so the two spellings cannot drift. The deprecation notice
goes to stderr; the desktop parses the URL off stdout, which is
unchanged.
Signed-off-by: Yi Lyu <isabellalyu1130@gmail.com>
Co-authored-by: Serena Ruan <82044803+serena-ruan@users.noreply.github.com>
Remove the cron schedule triggers from both discord-watch-rotation
workflows so they no longer fire automatically. workflow_dispatch is
kept for manual runs, and the original crons are left commented out so
the schedules can be restored later.
Co-authored-by: Isaac
Switching from a Codex session to a Claude Code session briefly painted
the Codex model (e.g. gpt-5.5) in the Claude session's composer before
correcting itself.
`switchTo` clears the session-scoped model fields but deliberately keeps
`selectedModel`, the cross-session sticky pick, so a CLI-created new chat
inherits the user's last choice. The native picker kind flips to Claude
immediately (the session query and sidebar row are already cached), so
for the whole snapshot round trip the composer resolved the sticky and
read the outgoing session's model.
Only surface the sticky once the session's own catalog vouches for it.
Pre-bind the catalog is empty, so the label waits instead of advertising
a model this session would reject; post-bind it is a no-op, since the
store only ever leaves a catalog-compatible sticky (or the override) in
`selectedModel`.
Co-authored-by: Isaac
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(server): file forked sessions into the source's project
Forking a session filed in a first-class project left the fork unfiled:
fork_conversation built the fork's metadata row without project_id. The
fork route now carries the source's project onto the fork, gated on the
forker owning that project (projects are owner-private, so a fork of a
shared session filed in someone else's project stays unfiled).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): refresh the project folder when a session is forked
A fork inherits the source's project, but the dialog only invalidated the
flat session list — each folder renders its own ["project-sessions", name]
query, which has no poll and converges only on an explicit invalidation.
The push stream can't cover it either: it skips the active session, and the
fork becomes active on navigate. So the clone stayed missing from its folder
until a reload or a re-navigation.
Adds an e2e regression test. It seeds the committed turn the fork action
anchors on straight into the store (new seed_committed_turn helper) instead
of driving a model turn, so it neither waits on nor inherits the flakiness
of the mock-LLM harness.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): stabilize reasoning indicators during active turns
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
A dropped host rendered two different indicators depending on incidental
state. The badge read the host tunnel directly (name + red dot), while
ChatPage passed a separate `hostOffline` prop derived from
`liveness.kind === "host_offline"` that replaced the name with generic
"Host is offline — click to reconnect" copy.
`host_offline` is far narrower than "the host tunnel is down": it also
requires the runner to be down (a live runner short-circuits to `online`),
the startup grace to have lapsed, and the host to be non-resumable. So the
same event — the host dropping — showed a passive, unclickable name when
the runner outlived the host, and a nameless reconnect prompt when it
didn't. The name is what tells the user which machine to go restart.
The badge now owns the decision: one shape (name + status dot) that turns
into a button opening the reconnect instructions whenever its bound host is
offline and reconnectable. A dormant resumable managed host stays passive —
the next message wakes it, so `omnigent host` would be wrong advice.
The reconnect dialog's state now comes from the session's host binding
rather than liveness, so a session whose runner outlived its host gets the
`omnigent host` command instead of the local `omnigent run --resume` one.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
Closes#866
The `omnigent` Homebrew formula has not built since 0.7.0, and the tap reported
green anyway, so 0.7.0, 0.8.0 and 0.8.1 all merged with no bottle — every user
compiled from source, and CEL policies silently did not work.
- **Root cause was the CEL migration.** #2970 swapped `cel-expr-python` for
`cel-python` on the premise that it is pure Python. It is not: `cel-python`
hard-depends on `google-re2`, whose sdist runs `bazel build` whenever
`GITHUB_ACTIONS` is set. The bazel dependency moved rather than disappeared.
- **Pin compiled extensions to upstream wheels.** `generate_formula.py` gains
`WHEEL_REQUIRED` / `PREFER_WHEEL` / `PURE_WHEEL` with abi3 and universal2
handling, so grpcio (by far the most expensive build), protobuf, regex,
uvloop, httptools, argon2-cffi-bindings, markupsafe, pyyaml, zstandard and
google-re2 stop being compiled. Native wheels rank above pure-Python ones, so
protobuf keeps its upb build instead of the slow fallback.
- **jiter, tiktoken and watchfiles keep building from source.** Their maturin
wheels carry no Mach-O install-name padding, so Homebrew relocation fails with
"Failed changing dylib ID" (#866). `pendulum` can go neither way — its wheel
cannot be relocated and its sdist does not link on 3.14 (pyo3 leaves
`_Py_NoneStruct` undefined) — so it takes the pure-Python wheel, which ships
no extension module at all.
- **A dropped dependency is now an error, not a warning.** A missing sdist used
to be skipped silently, yielding a formula whose venv lacked an import;
`--allow-no-sdist` is the explicit waiver. The formula test also asserts
`import re2, celpy`, since omnigent imports celpy behind `try/except
ImportError` and would otherwise disable policies silently.
- **Delete `update-homebrew.yml`.** It raced `homebrew-tap-pr.yml` on the same
`release: published` event and asserted on hand-maintained stanzas the
template no longer emits, so it failed on every run. Its one worthwhile part
moves into `homebrew-tap-pr.yml`: an admin/maintain gate on manual dispatch
(it writes to another repo with an App token), plus
`persist-credentials: false`. Its nightly `schedule` is deliberately NOT
carried over -- that cron only existed because `brew
update-python-resources` resolves through pip's `--uploaded-prior-to=P1D`
window and so could never see a same-day release. The generator runs `uv pip
compile --no-config` straight against PyPI, so the blindness it worked around
no longer exists, and a nightly regeneration would just burn a runner to
print "nothing to do".
Verified by building the generated formula in the tap, not by inspection.
- `omnigent-ai/homebrew-tap#18` contains **verbatim output of this
`generate_formula.py`** and bottled successfully on macos-15 and macos-26
(run 30944428771, `bottles_macos-15` / `bottles_macos-26` ≈ 37 MB each). This
is the check that matters: it proves the generator — not a hand-edit —
produces a buildable formula, so the next release regenerates something that
works.
- `omnigent-ai/homebrew-tap#17` carries the same fix for the shipped 0.8.1
formula and is green on all three runners, with `brew test` running
`import re2, celpy`. Inspected the bottle: `celpy/__init__.py`,
`re2/_re2.cpython-314-darwin.so`, and a relocated
`jiter/jiter.cpython-314-darwin.so`.
- Audited every pinned wheel by replaying Homebrew's own operation,
`install_name_tool -id <Cellar path>` against each extracted `.so`, so the
wheel/source split is evidence-based rather than guessed.
- `python3.12 -m py_compile`, `ruff check`, `ruff format --check`, `brew style`
(no offenses), `ruby -c`, plus stubbed-PyPI unit checks of the new failure
paths (missing sdist is fatal, `--allow-no-sdist` waives it, abi3 accepted,
free-threaded `cp314t` rejected).
- Confirmed generator output matches the green formula: same 100 resources,
identical sdist/wheel split, no non-comment differences.
N/A — release tooling, no user-visible UI.
- [x] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
The generator has no test suite in this repo, and its real contract — "the
emitted formula builds under Homebrew on macOS" — cannot be asserted here. It is
covered instead by building the generated formula on the tap's `brew test-bot`
matrix (homebrew-tap#18, bottles produced on macos-15 and macos-26). The two new
generator failure paths were exercised locally against stubbed PyPI metadata,
and every wheel pin was verified relocatable with `install_name_tool`.
`brew install omnigent` works again, and installs prebuilt wheels instead of
compiling grpcio and friends from source.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
The boxlite SDK's BoxOptions already supports disk_size_gb, but the
omnigent wrapper never threaded it through — every box got the SDK's
own default disk size with no way to override it. Add
sandbox.boxlite.disk_size_gb to the server config, alongside the
existing image/env knobs.
Signed-off-by: Avri Chen-Roth <11185446+the-mentor@users.noreply.github.com>
Co-authored-by: Avri Chen-Roth <11185446+the-mentor@users.noreply.github.com>
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
* refactor(web): decouple typography from interface geometry
Make the desktop font preference drive semantic text tokens while keeping icons, controls, and spacing fixed.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* refactor(web): migrate interface body text to text-ui
Use the settings-controlled semantic body token across shared components and application pages for consistent sizing.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* style(web): refine sidebar typography and empty states
Align sidebar hierarchy with settings-controlled tokens and make empty projects easier to scan.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* test(e2e-ui): align font size checks with typography tokens
Update browser assertions for the discrete desktop font token and its current bounds.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
* test(ui-snapshot): update typography visual baselines
Adopt the CI-rendered snapshots for the intentional settings-driven typography changes.
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
---------
Signed-off-by: Ajay Alfred <ajayalfred07@gmail.com>
`HostRegistry.deregister` was a bare `dict.pop`, and the host tunnel's receive
loop refreshed `conn.last_frame_at` without checking whether its connection was
still the registered one. The runner side already guards both (
`TunnelRegistry.deregister` takes a session guard and `mark_frame_seen` rejects a
superseded session); the host side did not, so a host could be left in a state
its own route handler never noticed.
Dropping a host registration from outside the route handler did not close the
socket or cancel its tasks. The ping loop kept writing `host_store.heartbeat`, so
the durable row stayed **online** while every `host_registry.get` reported the
host offline. Anything that resolves liveness from that row then waits for a
reconnect the host was never told to make, because from the host's side nothing
happened. `register` already poisons a replaced connection's outbound queue for
exactly this reason; `deregister` now does the same.
Three changes:
- `deregister` queues the `None` sentinel so the sender loop exits and the
socket tears down, letting the host redial.
- `deregister` takes an optional `conn` generation guard and returns whether it
removed an entry. The tunnel route gates its `set_offline` write on that
return, so a superseded handler reaching cleanup after a reconnect replaced it
can no longer evict the live connection or mark a live host offline.
- `mark_frame_seen` mirrors `TunnelRegistry.mark_frame_seen`: a frame only
refreshes liveness while its connection is current, and the receive loop stops
when it is not.
Six tests added to `tests/server/test_host_registry.py`; five of them fail
against the previous behavior.
Co-authored-by: Isaac
Signed-off-by: Mark Tai <mark.tai@databricks.com>
Co-authored-by: Mark Tai <mark.tai@databricks.com>
* feat(web): fold settled turns behind a 'Worked for Xs' row
Once a turn completes, the chat view collapses its whole process trace
(interstitial narration, tool-run folds, resolved approval cards,
reasoning) behind one muted 'Worked for Xs' expander with a hairline
rule, leaving only the final answer visible - mirroring the Codex
desktop treatment so it's obvious where reading starts instead of a
wall of uniform prose. Expanding the row replays the trace inline.
- Live turns keep their trace expanded; liveness comes from the
bubble's own lifecycle, not session status, so a completed turn
folds even while a later turn streams (and vice versa).
- partitionTurn splits a settled turn into foldable process, exempt
always-visible cards (pending elicitations, persistent
dispatch/routing cards, in-progress spinners), and the trailing
final answer; a turn with no trailing answer (interrupted / failed
/ tool-only) never folds. Resolved approval cards fold with the
trace in document order. Codex's trailing turn_diff bookkeeping
folds as process instead of masquerading as the answer.
- The 'Worked for Xs' duration spans the live stream clock while
streaming, or the items' server created_at stamps on reload;
ConversationItem.to_api_dict() now exposes created_at (additive)
to make the reload path possible.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* test(stores): expect created_at in the item API-shape round-trip
to_api_dict() now serializes created_at, so the exact-shape assertion
gains the store-assigned stamp.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* docs(web): demo screenshots + cross-clock note for the turn fold
Adds the collapsed/expanded 'Worked for Xs' screenshots referenced by
the PR description, documents that turnWorkedForS's first block picks
the clock branch, and pins the reverse mixed-clock direction
(live-first, epoch-last) as undefined.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): settle the turn lifecycle on bare terminal status edges
The 'Worked for Xs' fold (and the Fork action) only appeared after
navigating away and back: the turn lifecycle finalized live ONLY on a
session.status edge carrying a matching response id, but most idle
publishes carry none (the PTY-activity relay, orchestration teardown).
So a native turn ending on a bare idle cleared 'Working…' while the
bubble stayed 'streaming' forever — settled state was only re-derived
from the snapshot on reload.
- session_status: any terminal edge (idle/failed/waiting) now
finalizes a still-streaming turn, id-matched or not; cancelled is
preserved. The stray running->idle pair the policy-deny
short-circuit publishes mid-turn is healed by
reviveStrayCompletedResponse: live deltas for the turn flip it back
to streaming, so the misread is a brief flicker, not a mid-turn
fold.
- Mid-turn first open: the initial session bind now reopens the
streaming lifecycle from the snapshot's activeResponseId (mirroring
reconnectStatusPatch), so a running session's live turn renders
expanded instead of prematurely folded.
- e2e: test_bare_idle_finalizes_turn_and_folds drives the exact event
sequence (running+id -> items -> bare idle) against a real server
and asserts the fold forms in place, no reload.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): fold turns split by a sub-agent await, and ease the collapse
Two gaps in the 'Worked for Xs' fold, both visible on a turn that
dispatches sub-agents.
Fold never formed. Dispatching sub-agents ENDS the parent turn — it
must yield to await their results — and the inbox wake starts a new
turn under a new response id carrying the answer. That splits one
logical turn across bubbles: the first holds narration + tool calls
and no answer, the second holds the answer and no work. The fold
required both halves in ONE bubble, so neither qualified and the
narration stayed spread out unfolded. buildBubbles now flags a bubble
whose turn continues in a later assistant bubble (scanning past the
runtime [System: ...] wake markers, stopping at a real user turn),
and such a bubble folds its whole trace despite carrying no answer.
The flag participates in bubblesEqual so the memoized bubble actually
re-renders when its continuation lands.
Collapse was abrupt. The settled render swapped a tall expanded trace
for a one-line row in a single frame, which read as a partial page
reload. The fold now MOUNTS OPEN when the turn settles on screen and
closes on the next frame, so the steps visibly fold into the summary
row; settled history still mounts closed (nothing to animate away).
The height animation lives in index.css because it needs Radix's
measured --radix-collapsible-content-height, and is disabled under
prefers-reduced-motion.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): remove the jolt at the start of the turn-fold collapse
The collapse read as two motions. Measuring the bubble height every
frame through a settle showed why: inserting the summary row and the
fold's own padding/border grew the bubble ~43px TALLER in one frame,
and only then did the 200ms collapse run — a jolt up, then a ramp
down.
- The summary row now grows in (grid-template-rows 0fr -> 1fr) over
the same beat instead of appearing at full height, so row expanding
and trace shrinking net one monotonic shrink.
- The animated element carries no padding or border of its own: any
chrome there is height that lands before the collapse starts, which
is exactly the jolt. Expanded spacing comes from the row's hairline
above and the message column's gap below.
- The fold also animates when it appears on an already-mounted bubble,
not only when the turn itself settles — a turn split by a sub-agent
await folds when its continuation lands, and that case was snapping
shut with no animation at all.
Measured on a live server, same turn shape both times: leading jolt
43px -> 10px, and both the plain and sub-agent-split cases now show a
single animated ramp instead of a jump followed by one.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): stop the turn fold oscillating on codex sessions
On codex the fold flipped collapsed/expanded repeatedly as a turn
streamed. Instrumenting a live turn showed why: the server recorded
that turn as ONE response, but the client showed five to seven
bubbles. A streamed narration renders as its own transient 'live:'
preview bubble until its authoritative item replaces it, and reasoning
bursts group separately, so bubbles appear and merge away on every
delta. Each appearance gave an earlier bubble 'a later assistant
bubble' and marked it continued, folding a fragment; the merge
unmarked it and unfolded it again. Two fragments folded mid-turn as
'Worked for 1s' / 'Worked' rows carrying only a reasoning burst.
- markContinuedTurns only runs between turns: while a response is
streaming the transcript is mid-restructure, so nothing is marked.
Marks are sticky, so a bubble that has folded never reopens when the
next turn starts streaming.
- A continued bubble must also have RUN something (a tool call in its
process) to fold. That is the shape the flag exists for — narration
plus tool calls, then a yield to await sub-agents — and it keeps a
narration- or reasoning-only fragment from folding into a lone
'Worked' row with nothing behind it.
Measured on live codex turns, same prompt shape: fragments folding
mid-turn 2 -> 0, and the only remaining fold is the real one at turn
end.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): never fold a bubble made only of streaming artifacts
Residual codex flicker: the fold still appeared and vanished mid-turn,
just less often. This path never involved the continued flag, which is
why the previous guards only reduced the frequency.
Codex splits an in-flight turn into fragment bubbles — a reasoning
burst (ctx.itemId is null until its item is finalized) plus a 'live:'
narration preview. Their synthetic response id never matches
activeResponse, so walkBubbles labels them 'completed', and a fragment
holding reasoning + text satisfied the ordinary process-plus-answer
rule and folded. When the authoritative item replaced the preview the
fragment merged away and its fold went with it.
A genuine turn always carries at least one server-assigned item id, so
a bubble whose items are ALL null-id or 'live:'-prefixed is a fragment
of the turn still arriving and never folds. LIVE_ITEM_PREFIX moves to
lib/blocks.ts so the renderer and the store share one definition.
Verified by assertion: before this change a reasoning + live-preview
bubble rendered a fold; now it renders expanded. Two frame-exact
recordings of the reported prompt (63k frames, with approvals) showed
no fold disappearing, so this was found by construction rather than by
reproducing it live.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): render a native turn as one bubble live, so it folds like it does on reload
Root cause of the codex fold flicker, found by comparing the same
conversation live vs reloaded: 4 assistant bubbles and NO fold live, 1
bubble folded after a reload. A native turn was being split into
fragment bubbles while streaming and merged back into one on reload,
so the two views disagreed. walkBubbles groups by response id, and
three kinds of block carried the wrong one:
- Live text previews were stamped with a synthetic 'live:<id>' as
their response id, so each streamed narration broke the run. They
now adopt the live turn's id (falling back to the synthetic id when
no turn is tracked, so a preview can't join an unrelated bubble).
- A native harness emits no response.created, so the reducer never
learned the turn id and stamped its own blocks (reasoning, streamed
text) with a stale or empty one. A 'running' status edge carrying a
turn id IS the native turn-start signal, so the reducer adopts it --
without sealing an already-open section, since codex opens reasoning
~2s BEFORE that edge lands and closing would split one thought in
two.
- Blocks emitted in that ~2s window still carry no id, so the store
attributes the trailing unattributed run to the turn when the edge
names it.
With one bubble per turn, the fold condition stops oscillating: it was
flipping because the fragment boundaries moved as previews appeared
and merged, so whichever fragment momentarily had the
process-plus-answer shape folded and then unfolded.
Also: a trailing reasoning item no longer blocks the fold. Codex opens
a reasoning section as the turn ends, landing it after the final
message; reasoning is process, never the answer, so it peels into the
trace like the turn_diff wrap-up already did.
Measured on the reported prompt (with approvals), same shape each
time: bubbles 4 -> 1, and fold transitions went from 'never appears
live' to exactly one 0->1 the instant the turn ends, with zero
decreases (no flicker).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): make bubble grouping and the turn fold robust to a mid-turn connect
The codex flicker survived the response-id stamping fixes because their
premise was fragile: they depend on the client CATCHING the one
'running' status edge that names the turn. A tab that connects (or
reconnects) mid-turn never sees it — SSE replays no status edges — so
reasoning blocks carry rid "" and live previews fall back to their
synthetic ids. Attaching a fresh client to the reporting user's live
session reproduced it exactly: the persisted turn was ONE response, but
the page rendered up to ELEVEN bubbles, five of which folded mid-turn,
including one fold flip back open.
Two structural fixes, replacing edge-dependence with invariants:
- walkBubbles no longer splits a bubble on ANONYMOUS response ids
("" or live:*): such blocks only ever come from the live stream of
the turn around them, so they join it, and a group that OPENED on
anonymous blocks adopts the first real id that arrives. One turn is
now one bubble regardless of which edges the client happened to see.
Bubbles also stop keying off transient live: preview ids, so the
authoritative-item swap no longer remounts the bubble.
- The LAST assistant bubble never folds while the session is running,
even when its lifecycle reads settled — a mid-turn connect misreads
the live turn as 'completed', and folding it collapsed and reopened
the trace as its tail alternated between text and tools. The
session's terminal status edge folds it, which is the natural moment
anyway. Earlier bubbles still fold as usual while a later turn runs.
Verified by attaching mid-turn to a live codex run of the reported
prompt (with approvals): before, 8+ bubbles with 5 mid-turn folds and
a fold flip; after, one bubble, expanded throughout, folding exactly
once when the turn ends.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): one bubble — and one 'Worked for' fold — per user turn
The reported flow (a codex step-wise/goal turn) rendered SEVEN
'Worked for Xs' folds under one user message: codex publishes a
distinct response id per STEP on its status edges while the items all
carry the thread id, so each step opened a new bubble, and every
settled fragment folded separately once the turn ended. The server had
persisted the whole thing as ONE response.
- walkBubbles now groups ONE bubble per user turn: a response-id
change between two assistant blocks with no user message between
them is a continuation (step-wise sub-turns, retries, pre-edge
blocks), not a new turn. The group tracks the LATEST real id so
lifecycle follows the live edge. Blocks stamped a distinct id ON
PURPOSE — deny/failure sentinels and REQUEST-phase elicitations —
still open their own bubble, in both directions.
- Fold appearance is debounced (500ms of held eligibility): a
step-wise turn's between-step idle edge, or a stray idle before its
revive, reads settled for a moment and would otherwise fold and
reopen the trace. Losing eligibility hides the fold immediately, and
settled history still mounts folded with no delay.
Tests that pinned per-response grouping modeled adjacent turns with no
user message between them; real streams separate turns with one (the
inbox wake marker in the sub-agent flow), so they now include it. The
reducer-driven reused-callId test keeps its no-cross-pollination
assertions within the merged bubble.
Verified live: a simulated 5-step turn (distinct per-step edge ids,
one thread id) renders one bubble with zero mid-run folds and exactly
one fold at the end, and a real codex approval run folds once, 0.5s
after the turn ends.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): don't fold a live turn's partial work on a mid-turn refresh
Refreshing while a turn was parked on an approval collapsed the
partial trace into a premature 'Worked for' row. Two holes let the
last-bubble fold suppression miss the live turn on reload:
- The parked elicitation forms its own trailing assistant bubble whose
card ChatPage floats to the page bottom, leaving the bubble
item-less (it renders null) — and that phantom was counted as the
'last assistant' bubble, handing the actual trace to the fold.
lastRenderableAssistantIndex now skips item-less bubbles.
- On a step-wise codex turn the snapshot's active_response_id names
the STEP id while the items carry the thread id, so on reload the
trace's lifecycle reads 'completed' even though the turn is parked.
A pending elicitation now suppresses the last bubble's fold
directly: a card awaiting the user proves the turn is in flight
regardless of what the lifecycle or session status read.
Verified live: reloading a session parked on a codex command approval
keeps the trace expanded with the card visible, and a simulated
mid-turn reload with the step/thread id mismatch stays expanded until
the terminal idle edge, then folds once.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): no 'Worked for' flash when a reload lands between turn steps
Approving an elicitation and refreshing flashed the fold: a step-wise
codex turn publishes an idle edge when each step completes, and a
reload landing in the between-step gap reads fully settled — status
idle, no pending card, trace ending in narration text — so the fold
mounted instantly (the settled-history fast path), then the next
step's running edge cancelled it. Reproduced deterministically: fold
at 0.27s, gone at 1.66s.
Nothing in that snapshot can distinguish the gap from a real turn end,
but the trace's AGE can say how ambiguous it is: items carry server
created_at stamps, so the bubble now records its newest item's time.
The last assistant bubble mounted over a JUST-active trace (newest
item < 15s old) holds its fold for 3s instead of showing it instantly
— long enough for the next step's running edge to cancel it, so the
gap reload never folds at all. A reload after a genuine turn end folds
once the hold elapses, and old history still mounts folded with no
delay.
Verified live against the simulated gap: reload-in-gap shows no fold
ever (was flash-then-hide), reload-after-real-end folds at ~3s, and
stale-history mounts fold instantly (unit-tested).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): don't pop a settled turn's fold open while the next turn spins up
Once a real user message follows the last assistant bubble, a running
status belongs to the reply-in-flight for that newer input, so the
settled bubble's 'Worked for' fold must not be suppressed. Closes the
opencode dip where the prior fold opened for seconds until the new
turn's first item mirrored through the TUI.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* feat(web): scroll the expanded 'Worked for' trace into view
Clicking the fold expands the trace above the reading position, and the
browser's scroll anchoring keeps the answer below it stationary — the
work opens off the top of the viewport and the click looks like a no-op.
On a user-initiated expand whose row+trace don't fit the scroller, snap
the fold row to the top (before paint) so the trace reads from its
beginning. Fits-on-screen expands and the programmatic mount-collapse
animation don't scroll.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* test(e2e): cover the 'Worked for' fold across native wire shapes
Four deterministic events-API tests: step-wise per-step status edges
fold once with no mid-run flicker; items that switch response id
mid-turn still yield one fold per user message; a mid-turn reload keeps
partial work expanded until the terminal edge; and a settled turn's
fold holds through a follow-up send's item-less gap.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): always snap the fold row on user expand
The fits-on-screen fast path never held in practice: on the last turn
the stick-to-bottom scroller treats the 200ms expand animation as
appended content and re-pins the bottom, and elsewhere native scroll
anchoring pins the answer below — either way the growing trace glides
the row off the top and the click looks like a no-op. Snap the row to
the scroller top on every user expand (the upward scroll also unpins
stick-to-bottom) and park overflow-anchor for the animation.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): make the fold's expand snap win against the bottom-lock
Clicking 'Worked for' while the view is pinned at the bottom (the
resting position on the last turn) did nothing: the expand animation
opens at height 0, so the snap clamps against a scroller with no room,
and stick-to-bottom's resize handler then rides the growth to the
bottom — programmatic scrolls never unpin it. User expands now open at
full height in one frame (no height animation), release the bottom-lock
via a null-safe ConversationScrollLockContext (same recipe as
JumpToTopButton), and then snap the row to the top.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): land the fold's snap below the chat top fade
The snap parked the row 8px below the scroller edge — inside
chat-scroll-fade's transparent band (opaque only from 80px), so the
'Worked for' label sat scrolled-to-top yet invisible. The row's
scroll-margin-top now lives next to the fade definition (88px, plus the
iOS inset variant) so the two can't desync.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): restore the session busy signal when a live delta revives a turn
A stray idle edge clears sessionStatus before the revive flips the
turn back to streaming, so shouldQueueSend saw an idle session and let
a mid-turn send bypass the queue (and the Working indicator stayed dark
until the next running edge). The delta that triggers the revive proves
the session is mid-turn — restore sessionStatus: 'running' with it.
Local send status stays untouched: cross-client and TUI-typed turns
have no local send in flight.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* docs(web): drop the turn-fold demo screenshots from the repo
The PR description references them by pinned commit SHA, so the binary
assets don't need to live in the tree.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
## Related issue
N/A
## Summary
- The Databricks AI gateway only serves Claude requests in coding-agent mode when the `x-databricks-use-coding-agent-mode: true` request header is present; omnigent's Claude launches to the gateway did not send it.
- `ClaudeSDKExecutor`'s Databricks gateway env (`_resolve_gateway_env`) and native-claude's ucode launch config now pass `ANTHROPIC_CUSTOM_HEADERS=x-databricks-use-coding-agent-mode: true`, which Claude Code forwards verbatim as request headers (this survives the thinking-display gateway shim, which forwards all request headers).
- Generic-provider gateway envs (non-Databricks `key`/`gateway` providers) deliberately do not receive the header.
## Test Plan
- `uv run pytest tests/test_claude_native.py tests/inner/test_claude_sdk_executor.py -q` — 283 passed.
- Updated the ucode env exact-equality assertion and gateway-env tests to assert the header; added `test_generic_provider_gateway_omits_databricks_header` to pin the Databricks-only scoping.
- `uv run ruff check` and `uv run ruff format --check` on the touched files — clean.
## Demo
N/A
## Type of change
- [x] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [ ] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
N/A
## Changelog
Claude sessions routed through the Databricks AI gateway now send the `x-databricks-use-coding-agent-mode` header the gateway requires
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
Moves aravind-segu from `owners` to `owners_paused` in the 12 areas they
owned, so PR reviewer assignment and issue triage stop routing to them.
Readers use only `owners`; the 2+ owner check counts paused owners, so no
backfill was needed and no area is left without an active owner.
`policies` is now down to a single active owner (TomeHirata).
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(codex-native): clear the MCP startup band once the model starts working
The web chat showed 'Starting MCP servers (3/4): <name>' underneath an
agent that was visibly already working, sometimes for minutes.
Codex delivers per-server startup edges only to the connection that owns
the thread, so the forwarder synthesizes the round and settles it when
the thread goes idle after a turn, or when a config-derived window
elapses. Both are late: a server that never reaches a terminal state
(e.g. a misconfigured command that never handshakes) keeps the band
pinned for the whole first turn, and the window stretches to the slowest
configured startup_timeout_sec plus grace (135s for a 120s budget).
Settle on the first model-produced turn item as well. Codex defers turn
EXECUTION until the startup round ends, so assistant-side output proves
the round is over while the turn is still running - the same invariant
the idle-edge settle already relies on, observed at the earliest point
it can be. The band now covers only the genuine pre-turn wait.
The turn's userMessage item is excluded, and only parent-thread events
count: a turn is ACCEPTED (thread flips active, user message
materializes) mid-startup, and a collab child's turn says nothing about
the parent's round.
Two adjacent fixes fall out: a mid-turn reload no longer re-shows the
stale band from the session snapshot, and hitting Stop during a first
turn no longer reports 'cancelled' for servers whose startup had in fact
finished. The failed-turn diagnostic that names still-pending servers is
unaffected - a failed turn/start produces no model output, so no settle
precedes it.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(codex-native): settle the MCP round once, and add before/after visuals
Addresses review feedback on the settle-on-model-output change:
- Settle at most once per forwarder connection. The round is seeded
once per connection and never on thread rotation, so once model
output settles it the outcome cannot change; without a guard every
later item in the session re-read the bridge file to reach the same
idempotent no-op. A state flag short-circuits them, and the new test
re-populates the map behind the flag so dropping the guard fails
rather than passing on idempotency alone.
- Add the before/after chat captures the review asked for, taken at the
same point in the turn (agent running 'sleep 40') against servers
built from the same web UI, differing only in this fix.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* chore(codex-native): drop the committed demo screenshots
The before/after captures don't need to live in the repo; the same
evidence is in the PR description as the sampled A/B table and the
runner-log timeline.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(cursor-native): merge Omnigent MCP bridge into existing mcp.json instead of overwriting
write_mcp_config() previously called build_mcp_config() which returned a
dict with only the Omnigent bridge MCP server, then wrote it wholesale to
.cursor/mcp.json. This destroyed any user-configured MCP servers.
Now read the existing mcp.json, merge the Omnigent entry into mcpServers
leaving other keys intact, and write back the merged config.
Fixes#3083
Signed-off-by: lg320531124 <155300404+lg320531124@users.noreply.github.com>
* fix(cursor-native): guard malformed mcp.json and cover the merge path
A hand-edited .cursor/mcp.json can hold any JSON shape. The merge read it
and indexed straight into it, so a list/null root or a non-dict mcpServers
raised AttributeError/TypeError and took down the session launch, where the
old overwrite-always code could not.
Discard non-dict shapes before merging, swap the try/except/pass for
contextlib.suppress (SIM105), and add tests for the merge path (user server
plus a sibling top-level key survive) and the malformed shapes.
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(cursor-native): type the mcp.json merge against JsonObject
main moved this module off typing.Any onto JsonObject (dict[str, object]),
so the merge's `dict[str, Any]` annotation broke ruff F821 and pyrefly
once rebased, and indexing the object-valued mcpServers failed bad-index.
Narrow the loaded JSON with isinstance into a local `servers` dict (the
pattern opencode_native_provider already uses) and bind it back into
`existing`, so the write lands through the alias and stays typed.
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: lg320531124 <155300404+lg320531124@users.noreply.github.com>
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Co-authored-by: Dhruv Gupta <dhruv.gupta@databricks.com>
The example Postgres URLs in these docstrings are placeholders, but gitleaks'
`postgres-connection-string` rule matches the `scheme://name:secret@host` shape
and can't tell a placeholder from a live DSN. That makes them permanent false
positives: they show up in GitGuardian digests, and the Databricks pre-push hook
re-flags them on every new-branch push, since pushing a new branch re-scans
commits already on main. Working around that means reaching for
SKIP_SECRET_SCAN, which is a habit worth not having.
Switching the examples to angle-bracket placeholders sidesteps the rule (`<` and
`>` fall outside its username/password character classes), and reads more
clearly as a placeholder besides.
Docstrings and comments only: with docstrings stripped, the AST of every touched
file is byte-identical to before. Test files are deliberately left alone: their
URLs are live inputs and expected values, and one case exists specifically to
prove percent-encoded credentials survive the prefix rewrite, so rewriting it
would defeat the test. Those remaining findings are best marked as false
positives in the scanner instead.
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
The private Databricks secure-release repo was named in 9 places: three
workflow header comments, the `release.yml` run summary, a design-doc table
row, and four direct links into the private repo's file tree from
`editors/vscode/PUBLISHING.md`. None of it resolves for anyone outside
Databricks.
`release.yml` printed the name into its run summary on every release. Public
run summaries are world-readable, so a repo variable would keep leaking it.
The summary now prints the full command with `<secure-release-repo>` as the
only placeholder, so a release manager still gets something to paste and fill
in, and points at the runbook for the value.
The rest is a straight substitution to "a Databricks-internal secure-release
repo". `PUBLISHING.md` keeps the build half and defers the repo name and
workflow paths to the runbook.
No behaviour change: no trigger, input, permission, or step logic is touched.
The only executable change is the summary `echo` block, verified by extracting
it from the YAML and running it.
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
`RELEASING.md` documents the whole release pipeline, including the private
Databricks secure-release repo, its workflow filenames, and its dispatch
inputs. A public reader can't act on any of that, so per the thread with Corey
and Rice it moves to `omnigent-internal` (`RELEASING.md`).
This deletes the file here and repoints the six inbound "see RELEASING.md"
pointers (4 workflows, the changelog script) at "the maintainer release
runbook", so nothing links to a path that no longer exists.
Scrubbing the private repo name from the workflow comments and
`editors/vscode/PUBLISHING.md` is a separate follow-up.
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(sessions): expose persisted activity heartbeat
Signed-off-by: Solaris-star <820622658@qq.com>
* docs: broaden updated_at wording to cover session metadata edits
Per review feedback: updated_at also advances on title renames
(including auto-titling), agent switches, and archive toggles — not
just conversation item appends. An orchestrator treating it as a pure
item-append heartbeat should know a mid-stall rename resets the clock.
Broadened the docstring in SessionResponse and the SDK Session class,
and re-ran scripts/dump_openapi.py so the OpenAPI description matches.
---------
Signed-off-by: Solaris-star <820622658@qq.com>
Two bugs in the Kimi Code (kimi) harness integration:
Bug 1 - Omnigent could never detect a completed kimi login. The KIMI_KEY
install spec had no file-based login detector and the setup overview row was
hardcoded to "Not configured"/warn whenever the CLI was installed, so a
successful `kimi login` always showed as not signed in.
Fix: add a subprocess-free detector `kimi_auth.kimi_login_detected()` that
returns True when `~/.kimi-code/credentials/kimi-code.json` exists and is
non-empty (the file `kimi login` writes; verified against kimi CLI v0.29.1),
mirroring the Gemini `gemini_login_detected()` pattern. Wire it into
`harness_readiness._FAMILY_CREDENTIAL_CHECK` (binary + credential gating, like
agy) and make the setup overview row render green "Signed in" when detected.
Bug 2 - Sign-out was broken. The spec declared `logout_args=("logout",)` but
kimi has no `logout` subcommand (`kimi logout` errors "unknown command" on
v0.29.1). Set `logout_args=None` so `harness_logout` is a no-op for kimi (same
as Qwen / agy) and remove the "Sign out (kimi logout)" row and its branch from
the Kimi drill-in. Docstrings/comments claiming kimi ships `kimi logout` are
corrected.
Tests: add tests/onboarding/test_kimi_auth.py (present/absent/empty credential
via tmp paths), update the harness_install/harness_readiness onboarding tests
for the new logout_args=None and binary+credential readiness, and update the
CLI drill-in / setup-overview tests (no sign-out row; signed-in vs
not-configured overview row).
Signed-off-by: evangoh122 <evangohsg@gmail.com>
Signed-off-by: Evan Goh <authoremail@example.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Approving a plan from the web UI did nothing: the card showed as
approved but the plan never ran, and answering in the terminal view was
the only way through. Claude Code ignores a PermissionRequest hook's
`allow` for ExitPlanMode (that dialog only accepts a TUI answer), so the
`setMode` decision the server builds never took effect. As a result
Claude's `auto` mode was unreachable from the web UI, since the plan
card is the only surface that offers it.
Key the verdict into the pane instead, the way a local user would:
option 1 for accept-with-auto-mode, 2 for accept, Escape for reject.
The bridge only presses a key when the plan dialog is actually on
screen, which keeps a non-plan verdict (or one already answered in the
terminal) a no-op. Rides the approval event the server already forwards
to the runner, so no new event type or server plumbing.
Co-authored-by: Isaac
Signed-off-by: Pranav Setlur <psetlur@gmail.com>
The Databricks Apps entrypoint built every other store but never the
project store, and create_app mounts the projects router only when a
project store is wired — so first-class Projects were non-functional
on every Databricks Apps deployment while the bundled web UI still
offered project creation. The CLI server and Docker entrypoint paths
already wire it.
Construct SqlAlchemyProjectStore from the Lakebase DB URI and pass it
to create_app, mirroring the other stores.
Co-authored-by: Isaac
Claude-Session: https://claude.ai/code/session_01P9dr2dYHrwMvnXvJsjLDKk
Signed-off-by: Randy 🌞 <randypitcherii@gmail.com>
* Central CTA + background bugfix
Landing screen:
- Headline moves to Hanken Grotesk at 400 weight ("What should we build?"),
self-hosted via @fontsource-variable so no CDN is involved, exposed as the
`font-display-alt` token.
- The project variant swaps the bare folder glyph for a pink rounded tile,
using a new `tag-pink` token from the design's tag palette.
- The composer placeholder and its aria-label now name the selected project
("Start a new session in <project>") instead of always reading the generic
task prompt.
Bug fix — the mobile sidebar was see-through. Below md the sidebar is a
full-screen overlay on top of the chat, but the per-theme canvas rules paint
it with the `background` shorthand, which resets background-color and silently
overrode Sidebar.tsx's max-md:bg-card-solid; the dark stack is entirely
translucent, so the conversation showed straight through. Restores an opaque
fill under the gradients below md only, at matching specificity and after the
theme rules, so desktop keeps its intended translucency.
Adds regression tests for that contract, and updates the landing-screen tests
and visual-suite docs for the new headline.
Co-authored-by: Isaac
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
* test(e2e-ui): regenerate visual baselines
---------
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
* perf(runner): bound per-runner memory via glibc arenas + threadpool cap
Each session spawns its own runner process, and each grows to ~200MB in
prod, over-using host resources. Profiling shows ~123MB is the irreducible
import floor; the growth on top is runtime bloat from threaded Python on
glibc: the runner offloads heavily via asyncio.to_thread, the default
executor sizes to min(32, cpu+4) threads, and glibc opens up to 8*ncpu
malloc arenas that never return to the OS. Nothing tuned any of this.
Three low-risk, env-gated levers (all no-ops or benign off Linux):
- MALLOC_ARENA_MAX=2 + a 128 MiB trim threshold, injected into the runner
child env at both spawn sites via a shared _proc.malloc_tuning_env()
helper. Empty off Linux; OMNIGENT_RUNNER_MALLOC_ARENA_MAX=0 reverts.
- Cap the asyncio default executor at 8 workers (runner.threadpool_max_workers
config key, OMNIGENT_RUNNER_THREADPOOL_MAX_WORKERS env override), set before
any to_thread use so the 20-thread default pool is never created.
- gc.freeze() after app construction to drop the static import graph from
GC's tracked set.
This targets the runtime growth, not the import floor; collapsing the floor
itself (a copy-on-write zygote) is tracked separately.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): apply the glibc arena cap at the zygote exec
The zygote forkserver landed and is now the default runner spawn path, which
silently defeated this branch's MALLOC_ARENA_MAX injection. glibc reads that
variable once, when its allocator initializes at exec; a zygote-forked runner
never execs, it just replaces os.environ, so the value arrived far too late to
configure an allocator and the cap stopped applying to every runner.
Move the injection to the zygote's own Popen -- the single real exec on this
path -- so all forked runners and harnesses inherit an already-capped
allocator. Two tests pin the contract at that boundary, including that an
operator's explicit export still wins.
The other two levers on this branch (the 8-worker threadpool cap and
gc.freeze()) live inside _run_tunnel_from_env, which every runner reaches
regardless of how it was started, so they were unaffected. Note in
malloc_tuning_env why the arena cap is glibc-only: macOS libmalloc uses
per-CPU magazines with madvise reclaim and ignores MALLOC_ARENA_MAX, so macOS
hosts get their reduction from the threadpool cap (measured: 21 threads -> 9).
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
nimble-python is the optional `nimble` extra and the import is already
guarded by try/except ImportError. Pyrefly has no way to know it's
intentionally absent, so annotate with `# pyrefly: ignore[missing-import]`
to silence the false-positive without changing runtime behaviour.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(policies): add force-push protection to GitHub policy
Add a `deny_force_push` parameter (default `True`) to the GitHub
policy that blocks `git push` with force flags (`--force`, `-f`,
`--force-with-lease`, `--force-if-includes`) regardless of
repo/branch allowlists. This prevents agents from rewriting remote
history, which can destroy commits and break collaborators' clones.
The check fires before repo/branch gating so even a force push to
an undeterminable remote alias is denied rather than surfaced as ASK.
Set `deny_force_push=False` to let force pushes through normal
write gating.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(policies): merge startswith calls to satisfy ruff PIE810
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* style(policies): join force-push condition onto one line for ruff format
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
---------
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* spike(runner): measure copy-on-write savings from a warm-fork zygote
Each session spawns its own runner process, and each pays a ~123MB import
floor for omnigent's own graph plus pydantic/fastapi/httpx. Runtime tuning
trims the growth on top but can't touch that floor; the only way to collapse
it is to import the graph once in a warm parent and os.fork() a child per
session, sharing the read-only import pages copy-on-write.
This standalone script measures whether that COW sharing actually
materializes before we commit to the full zygote architecture. It imports the
runner graph once, forks N idle children, and reports aggregate memory against
an N-process Popen baseline, optionally with gc.freeze().
Not wired into the daemon — this is a measurement gate, not a feature. On this
macOS box (N=8) the fork path showed ~82% lower aggregate footprint than the
Popen baseline, but macOS phys_footprint is only an indicative analog to Linux
Pss and the children idle (no COW erosion from refcount page-dirtying), so a
Linux-under-load measurement is still required before productionizing.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(runner): add copy-on-write zygote forkserver for runner processes
Every session spawns its own runner, and each pays the full ~120MB import
floor (omnigent's graph + pydantic/fastapi/httpx). On a host running N
sessions that floor is duplicated N times. This adds a zygote: a single
long-lived process that imports the runner graph once and os.fork()s a child
per session, so on Linux the read-only import pages are shared copy-on-write
and each extra runner costs only the pages it dirties.
Design (grounded in the daemon/runner lifecycle, not the naive sketch):
- omnigent/runner/_zygote.py — the forkserver. Single-threaded, no event loop
or network; imports the graph once, gc.freeze()s it, then blocks on an
AF_UNIX control socket forking a child per request. The child reopens its
log, replaces os.environ with the request env, and calls the unchanged
_entry.main() — so it behaves exactly like `python -m omnigent.runner._entry`.
It is Popen-exec'd by the daemon (never forked from it), so it inherits none
of the daemon's asyncio loop / websocket / worker threads — the classic
fork-in-multithreaded-async deadlock is avoided by construction.
- omnigent/host/runner_zygote.py — the daemon-side client. ZygoteManager owns
the control socket; ZygoteRunnerProc is a Popen-shaped shim so the existing
_RunnerHandle / _watch_runner / _handle_stop paths are unchanged. The daemon
is NOT the forked runner's parent, so poll()/returncode/wait() round-trip to
the zygote (the real parent) for exit status while terminate()/kill() signal
the pid directly.
- connect.py — _handle_launch forks via the zygote when enabled, else the
original Popen. RUNNER_PARENT_PID is set to the ZYGOTE's pid (not the
daemon's) because the runner's orphan watchdog compares os.getppid(); daemon
death -> control-socket EOF -> zygote exit -> runners reparent -> each tears
itself down, preserving today's parent-death semantics through one hop.
Gated behind OMNIGENT_RUNNER_ZYGOTE=1 and Linux-only; any zygote failure
disables it for the daemon's life and falls back to a direct Popen, so it is
never a hard dependency. Also removes the Phase-1 measurement spike script,
which this supersedes.
Verified on macOS: a real zygote subprocess forks children, reports pids and
exit codes, isolates per-fork env, reaps cleanly, and tears down on stop
(fork works on macOS even though the COW savings are Linux-only). The
production memory win and a full session-through-the-tunnel run are unverified
here — they need a Linux host under load, which this change is written to be
turned on for.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): address zygote review feedback
- connect.py: a failed fork no longer stops the running zygote. Stopping it
would kill healthy runners already forked from it (their orphan watchdog
sees the parent die), so one bad fork could take down unrelated live
sessions. Latch a `_zygote_disabled` flag for future launches instead and
retain the manager so the zygote is still reaped on daemon shutdown.
- runner_zygote.py: wait() after kill() in stop() so a zygote that ignored
SIGTERM is reaped rather than lingering as a zombie.
- _zygote.py: unify the _entry/app/native import to a single `from ... import`
(CodeQL flagged mixed import styles).
- test: build the fresh-interpreter probe via an explicit newline join instead
of implicit adjacent-string concatenation (CodeQL).
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): forward --log-to-stderr TTY fd through the zygote
The direct-Popen launch path forwards OMNIGENT_LOG_TTY_FD via
child_logging_popen_kwargs so a detached runner can still mirror logs to the
daemon's terminal. The zygote path dropped it, so --log-to-stderr mirroring
was lost for zygote-forked runners.
Forward it across both hops:
- daemon -> zygote: reuse child_logging_popen_kwargs to dup the TTY fd and add
it to the zygote's pass_fds (the helper also rewrites env[LOG_TTY_FD] to the
duped number).
- zygote -> forked runner: the valid fd number inside the child is the one the
zygote inherited, not the daemon-side number the payload carries, so the
child restores LOG_TTY_FD from the zygote's own value (and clears a stale
payload value when the zygote has no terminal mirror).
Adds a test asserting a bogus payload LOG_TTY_FD is cleared in the child.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): address second round of zygote review feedback
- _zygote.py: create the forked child's log file 0o600, not 0o644. Runner
logs can carry secrets (tokens, prompts); matches create_process_log_path.
- _zygote.py: the child guard now preserves SystemExit's code instead of
flattening it to a traceback + exit 1, so a zygote-forked runner exits with
the same code as `python -m omnigent.runner._entry` (main() raises
SystemExit on a tunnel rejection). New test covers it via a raise seam.
- runner_zygote.py: stop the partially-started zygote if the initial ping
raises (timeout / EOF), so a failed start never leaks a process + socket.
- runner_zygote.py: signal via signal.SIGTERM / signal.SIGKILL instead of the
raw 15 / 9.
- test: mark the suite posix_only (it uses os.fork / pass_fds) so cross-
platform sweeps skip it on Windows.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(runner): enable the zygote on all POSIX hosts, not just Linux
The host daemon runs on the user's own machine — most often macOS — so a
Linux-only gate denied the copy-on-write import-floor savings to the majority
of hosts. Gate on IS_POSIX instead (the zygote needs os.fork + AF_UNIX
fd-passing, both POSIX; Windows still takes the direct Popen path).
macOS is the platform where fork-without-exec is riskiest (CoreFoundation/GCD
abort a forked child that touches them), so this was verified rather than
assumed. The abort is triggered by forking from a MULTI-threaded process, which
the zygote already designs against: it forks from a single-threaded parent
(asserted active_count()==1) and does create_app + all network work in the
child. Evidence on this macOS box:
- A faithful fork probe (fork from the single-threaded import state, child runs
create_app + getaddrinfo + TLS ctx + asyncio + httpx) survived 5/5. The same
work forked from a multi-threaded parent SIGSEGV'd 2/3 — confirming the
single-threaded fork is what makes it safe.
- test_host_launch_runner_and_session_round_trip passes with
OMNIGENT_RUNNER_ZYGOTE=1: a real host daemon forks a runner through the
zygote, the runner connects its tunnel, and a full mock-LLM session round-trip
completes. The daemon log confirms the zygote path (distinct zygote/runner
pids), not a Popen fallback.
Also adds an info log on the successful zygote-fork path so operators can see
the zygote is active and which pids are involved.
Still opt-in behind OMNIGENT_RUNNER_ZYGOTE=1 with the full Popen fallback; the
steady-state Pss win under load remains best measured on a Linux host.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(runner): fork harness subprocesses from the runner zygote
The harness subprocess (`python -m omnigent.runtime.harnesses._runner`) is a
separate exec per conversation, so it re-pays its import floor — and that floor
is ~54MB of the same common graph (fastapi/pydantic/omnigent-core) the runner
zygote already holds resident. This extends the zygote to fork harness children
too, sharing that graph copy-on-write instead of exec'ing a fresh interpreter.
- _zygote.py: the serve loop becomes a single-threaded `selectors` multiplexer
over the daemon socket PLUS one inherited control socket per forked runner.
A new `fork_harness` command forks a child that reproduces `_runner.main(argv)`
in-process. The runner-fork request/response bytes are unchanged; the new
multiplexer wraps them rather than rewriting them. A forked child closes every
inherited zygote-side socket (it never speaks the fork protocol).
- _harness_zygote_client.py (new): the runner-side client. `HarnessZygoteClient`
reads the inherited control-socket fd from OMNIGENT_RUNNER_ZYGOTE_HARNESS_FD;
`ZygoteHarnessProc` is an asyncio.subprocess.Process-shaped shim (pid /
returncode / wait / send_signal / kill) with a background poll task keeping
returncode fresh for _wait_for_bind's synchronous reads.
- process_manager.py: `_spawn_harness_process` forks via the zygote when the
runner was itself zygote-forked, else the original create_subprocess_exec;
disabled on first failure so it falls back for the process's life.
- _runner.py: a zygote-forked harness has the zygote (not the runner) as OS
parent, so its watchdog probes the runner pid explicitly instead of trusting
os.getppid(), and skips PR_SET_PDEATHSIG (which would bind death to the
zygote). Gated by OMNIGENT_HARNESS_ZYGOTE_FORKED.
Present only when the runner itself was zygote-forked; any failure falls back to
a direct exec, so the harness fork is never a hard dependency. The win is
bounded to the ~54MB Python wrapper (the external claude/codex CLI is a separate
exec no Python zygote can share) and materializes under multi-conversation
fan-out. Verified on macOS: fork_harness forks, reports pid + exit code,
round-trips argv, reaps, and leaves the daemon socket serving; existing
process_manager tests unchanged. Linux Pss savings still unverified.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): clear pyrefly type errors in the zygote
- ZygoteRunnerProc.wait: narrow on `timeout` (not just `deadline`) so
TimeoutExpired(timeout=...) gets a `float`, not `float | None`.
- _spawn_zygote_process: pass stdin/stdout/stderr explicitly with a typed
`BinaryIO | None` log handle instead of a `dict[str, object]` splat that
matched no Popen overload.
- _ZygoteServer.serve: cast selector key.fileobj (HasFileno | int) to socket
— only sockets are ever registered.
- _ZygoteServer._on_readable: wrap the bytearray partition result in bytes()
before dispatch, which expects bytes.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): harden zygote failure paths (crash recovery, exit-code leak)
Review flagged three correctness bugs in the unhappy lifecycle paths; none are
security issues but each is reachable in prod.
1. Unexpected zygote crash stranded the daemon's view of every child. The
daemon isn't the runner's OS parent, so once the zygote died it had no
channel to learn a runner exited — ZygoteManager.poll returned None
("still live") forever, so _watch_runner looped, _handle_runner_status
reported gone sessions as alive, and _handle_stop's final wait() could hang.
Now poll() probes the runner pid directly when the zygote is gone: a dead
pid surfaces a non-zero sentinel (254) so the runner reads as dead-and-
failed, not eternal alive. _handle_stop's post-kill wait() is now bounded.
2. _exit_codes leaked for a dropped runner's harness children. Exit codes were
only popped via poll, but a dropped runner's harnesses have no remaining
client to poll them — the entries accumulated (unbounded map growth +
pid-reuse misattribution). _drop_runner now discards those descendants'
codes and marks still-live ones orphaned: _reap waitpid's them (no zombies)
but discards the code instead of storing it.
3. ZygoteHarnessProc.wait() masked a crashed harness as exit 0. If the zygote
went away, wait() returned 0, so a harness that crashed on boot (bind
failure, import error) read as a clean exit and the process manager could
hang waiting for a bind that never comes. Now probes the harness pid and
returns a non-zero sentinel when the code is unrecoverable.
Also: tighten "Linux-only" docstrings to "POSIX; COW savings on Linux" (the
gate is IS_POSIX and the path runs on macOS), and add a sleep test-seam so the
new failure-path tests can hold a child genuinely alive.
Tests: kill the zygote under a live runner and assert the daemon eventually
sees it dead (not hanging); a dropped runner's harness code is not retained; a
crashed harness with an unrecoverable code surfaces as failure, not 0.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): keep zygote poll/wait off the daemon event loop
Review flagged a liveness regression on the enabled path: for a zygote-forked
runner, poll()/wait() are blocking control-socket round-trips (with lock
contention against a booting zygote that holds the lock across its ~120MB
import), not the lock-free waitpid the direct-Popen path used. Calling them on
the loop thread could freeze the whole daemon — all sessions, websocket
traffic, heartbeats — until the import finishes or the 30s control timeout
elapses.
- _watch_runner: poll() now runs via asyncio.to_thread.
- _handle_stop: now async; the poll/terminate/wait sequence runs off-loop in a
_stop_runner_proc helper. Its dispatch site and three tests updated to await.
- _tracked_runner_pids: include the zygote pid so the orphan reaper never
waitpid's the zygote out from under ZygoteManager._proc on an unexpected
crash (which would confuse is_running()/stop()).
Also updates test_poll_after_stop to use a live child, since the crash-recovery
sentinel (254) now correctly fires for an already-exited pid after stop().
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): status query off-loop + enable zygote by default
- _handle_runner_status did its poll() on the event loop, the one place the
PR hadn't moved off it. For a zygote-forked runner poll() is a blocking
control-socket round-trip (bounded only by the 30s control timeout, and
contended against a booting zygote), so a slow zygote could stall the whole
daemon for a single status query. Made it async and run the poll via
asyncio.to_thread, matching _watch_runner / _handle_stop. Dispatch site and
the three status tests updated to await.
- Enable the zygote by default: OMNIGENT_RUNNER_ZYGOTE is now opt-OUT
(=0/false/no/off), not opt-in. The host daemon runs on the user's own
machine (most often macOS), so defaulting on lets most users share the
~120MB import floor. Still POSIX-gated with a full Popen fallback, so an
unsupported platform or any zygote failure is transparent.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: prevent mid-spawn launch leaks and harden zygote request handling
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(web): set the text size steps from the design
Body and chat-thread text are both 13px/18px in the design; the shared
`text-13` step was on a 20px line, so tighten it to 18px. Adds the 12px/16px
caption step used by sidebar section subtitles (Projects, Sessions).
Defines the steps only — switching each surface onto them is follow-up work.
Co-authored-by: Isaac
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
* feat(web): put chat and sidebar text on the design's type scale
The chat thread hard-coded its own 15px/24px with negative tracking, and
sidebar rows set a size but no line height, so neither matched the design.
- Chat bubbles (user and assistant share the wrapper): 13px/18px, and the
-0.01em tracking is dropped — the design specifies 0.
- Sidebar body rows: pin the line height to 18/13 of the font size, which was
previously left to inherit.
Both stay in rem/unitless so the mobile root-font bump and the Appearance
font-size setting keep scaling them. Sidebar section captions were already
12px/16px and are unchanged.
Co-authored-by: Isaac
* test(e2e-ui): regenerate visual baselines
* refactor(web): express the sidebar line height in rem
1.3846 was the 18/13 ratio written as a unitless number — unreadable, and it
took arithmetic to confirm it meant 18px. 1.125rem is 18px directly and
scales the same way, matching how the chat wrapper states it.
Co-authored-by: Isaac
---------
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
OMNIGENT_RUNNER_ENV_PASSTHROUGH lets an operator name extra env vars for the
host to forward on to spawned runners (provider gateway wiring, config env: refs,
etc.). It worked locally but was a silent no-op in --server mode: the remote
daemon env is allowlisted by a prefix set of DATABRICKS_ + LC_/MLFLOW_/OTEL_/
OMNIGENT_OTEL_ — NOT plain OMNIGENT_ — so the control var itself was stripped at
the CLI->daemon hop, and _build_runner_env never saw the names it listed. Any var
forwarded through the passthrough (e.g. a Linear API key for the repro-agent)
reached the runner locally but never remotely.
Add OMNIGENT_RUNNER_ENV_PASSTHROUGH to _RUNNER_ENV_ALLOWLIST so it survives both
hops. It carries only env var NAMES, not secrets, so allowlisting it leaks
nothing on its own — each named var must still independently reach the daemon
(here via the DATABRICKS_ prefix).
Tests: a daemon-hop test (remote env keeps the control var) and an end-to-end
two-hop test (a named var survives CLI->daemon->runner, an unnamed one doesn't).
Both fail without the one-line allowlist change.
Co-authored-by: Isaac
* dev/repro-agent: give Linear tickets a real fetch path + follow linked GitHub issues
The local repro-agent pointed Linear tickets at nonexistent "Linear tools",
so Linear runs had no way to read the ticket body and fell back to guessing
from the URL slug — noticeably worse reproductions than GitHub issues, which
have a working `gh issue view` path.
Wire Linear to the same GraphQL path the internal issue-sync agent uses
(api.linear.app/graphql, `Authorization: $LINEAR_API_KEY`, no Bearer), pulling
description/comments/attachments via sys_os_shell. When the key is absent or
auth fails, stop with needs_more_info naming the missing key instead of
guessing. Also: when a Linear ticket links a GitHub issue, always fetch that
issue too and treat it as authoritative for the technical journey — that
richer thread is why GitHub-first runs reproduced better.
Co-authored-by: Isaac
* dev/repro: forward the Linear key through the --server env strip
Reading a Linear ticket needs the key in the agent's shell, but under --server
the CLI->daemon->runner hops strip everything not allowlisted. The DATABRICKS_
prefix survives only the first hop; the daemon->runner hop has no DATABRICKS_
prefix. So dev/repro.py now names DATABRICKS_LINEAR_API_KEY in
OMNIGENT_RUNNER_ENV_PASSTHROUGH (itself allowlisted) when a Linear URL is passed
and the key is set, which forwards it the rest of the way. AGENTS.md reads
whichever name is present (LINEAR_API_KEY locally, DATABRICKS_LINEAR_API_KEY
under --server). Warns rather than fails when the key is missing.
Companion change (omnigent-internal): the repro-agent CI workflow must set
DATABRICKS_LINEAR_API_KEY from secrets.LINEAR_API_KEY in the run step, mirroring
how it already sets DATABRICKS_BEARER for the LLM key.
Co-authored-by: Isaac
* dev/repro: mirror LINEAR_API_KEY into the DATABRICKS_ name
Maintainers typically export the plain LINEAR_API_KEY locally, so copy it into
DATABRICKS_LINEAR_API_KEY when only the plain name is set — then the same
passthrough forwarding carries it past the --server env strip. Warn only when
neither is set.
Co-authored-by: Isaac
* feat(web): make the rails flush boxes and move the canvas gradient
The sidebar and workspace rails were floating cards (margin, rounded
corners, border, shadow) on a gradient canvas. The design has them flush to
the window edges, reading as part of the canvas.
- Left sidebar and right workspace rail sit flush: no outer margin, no
rounding, no drop shadow. The workspace rail keeps a left divider.
- Light canvas is flat white; the brand gradient moves onto the left
sidebar, joined by the mock's dot-grid and pink corner glow.
- Dark canvas carries the mock's purple gradient; the dark sidebar gets the
same dot-grid plus a purple bottom wash and the diagonal sheen.
- Both rails are excluded from the dark glass rule instead of overriding it,
so they no longer pick up its blur, sheen, fill, or border. The workspace
rail's panel contents are transparent too.
- Dark surface tokens (--card, --card-solid, --tray, --muted, --background)
move off their purple tint onto neutral slate.
Consolidates the canvas/rail CSS so each surface owns its full background in
one rule, and drops the now-redundant ::before dot overlay.
Co-authored-by: Isaac
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
* test(e2e-ui): regenerate visual baselines
---------
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
Broaden issue-triage auto-assignment from P0/P1-only to every triaged
issue except needs_info ones. The gate now keys off needs_info alone, so
any bug/enhancement/doc issue with enough info to triage gets a
load-balanced area owner (least open assigned issues first, LLM rank as
tiebreaker) instead of only high-priority ones. Drops the now-unused
priority/type branch from the shell gate.
Also refresh .github/areas.json ownership:
- remove SabhyaC26 from all areas
- add PattaraS to harness-antigravity (keeps it at the 2-owner minimum)
- reactivate dbczumar (owners_paused -> owners) across their areas
Co-authored-by: Isaac
* fix(web): remount terminal view when switching same-vendor sessions
Two sessions of the same shape share a fixed agent-terminal id (e.g. every
claude-native session's `terminal_claude_main`, every SDK session's
`terminal_tui_main`). ChatPage stays mounted across a session switch and only
feeds MainTerminalView / TerminalsPanel a new conversationId, so keying the
xterm wrapper on the terminal id alone let React reuse the existing mount —
the pane kept the previous session's 20k-line scrollback until the new
WebSocket reconnected and tmux repainted. The stale history cleared only on a
manual refresh.
Scope the wrapper key to `${conversationId}:${terminalId}` in both surfaces so
a session switch forces a clean remount (fresh xterm + WebSocket, no stale
buffer). Add regression tests that switching conversationId with the same
terminal id remounts the TerminalView.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(web): assign terminal mount id once per mount
Copilot review flagged that useRef(++terminalMountSeq) evaluates the
increment on every render (useRef ignores the arg after first render),
so the module counter advanced on re-renders — contradicting the
comment. The read value (instance.current) was still stable, so the
assertion held, but assign the id conditionally so the counter tracks
real mounts.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
The Chat/Terminal switcher moved into the header as a Radix DropdownMenu
whose trigger toggles on pointer-down and carries a controlled hover
tooltip on the same node (ViewModeToggle.tsx). On a busy page — a live
terminal stream plus that tooltip re-rendering during the click — a lone
`.click()` occasionally nets the menu back to closed, so the follow-up
`expect(menuitemradio).to_be_visible()` times out. That is the observed
flake in test_codex_goal_mode and the native render-parity suites: the
failure snapshot shows `tooltip "Terminal view"` (rendered only while the
menu is closed) with no menu items.
Add a shared `_select_view_mode(page, option)` helper that reopens the
menu in a retry loop until the target radio item is actually visible, then
selects it, instead of trusting a single toggle click. Route
`_ensure_chat_view` and every native-parity `_open_terminal_view`
(codex, claude, goose, hermes, cursor, kiro) through it.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Repoint the gray text and border tokens onto the shadcn Zinc scale so the
UI's neutrals match the design system:
- Primary text (--foreground, --card-foreground, --secondary-foreground,
--sidebar-foreground) -> Zinc 800 #27272a
- Secondary text (--muted-foreground) -> Zinc 500 #71717a
- Default border (--border, --input, --sidebar-border) -> Zinc 200 #e4e4e7
- Strong border (--border-strong) -> Zinc 400 #a1a1aa
Also adds the two tokens the palette needs but the app lacked:
--border-weak (Zinc 150) and --foreground-tertiary (Zinc 400), exposed as
Tailwind utilities.
Co-authored-by: Isaac
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: Hubert Zub <hubert.zub@databricks.com>
The codex goal-mode + native-parity targets need the Codex-parity Rust
sidecar. flake-stress-ui.yml relied on the fixture's inline `cargo build`
at test time, capped by --timeout=300. On a cold Rust cache every parallel
attempt independently compiles the ~1100-crate tree and overruns the
per-test timeout, so all attempts die at fixture setup before the test
body ever runs — masquerading as a 100% failure rate unrelated to the
target under test.
Mirror e2e-ui.yml / ci.yml: add a dedicated build-sidecar job that
compiles the sidecar once (same main-scoped cache key so it usually
restores), uploads the ~10MB binary, and has each attempt download it and
set CODEX_PARITY_SIDECAR_BIN. build_sidecar_bin() then returns the prebuilt
path and skips cargo entirely. Drops the per-attempt Rust toolchain + target
-dir cache that never made the inline build fit the timeout.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
`dev/repro.py --public` sets `public: true` in the agent's input contract, and
the agent shares the session read-only (`sys_session_share __public__`) at the
start of its run so it is browsable live — useful when watching a run or
reproducing against a shared --server. Off by default (a local session is
already yours to browse).
- dev/repro.py: add --public; include `"public": true` in the payload when set.
- config.yaml: re-add `agent_session_sharing: public` to grant the __public__
capability (opt-in via the flag).
- AGENTS.md: document the `public` input; make sharing the first preflight step.
- README.md: document the --public flag.
Co-authored-by: Isaac
* dev/repro: add worktree-isolating driver script; clarify browser context
Add dev/repro.py — a maintainer-only wrapper around `omnigent run
dev/repro-agent`. It prompts for the bug URL (or takes it as an argument /
bare id like OMNI-1234 / 3987), creates an isolated git worktree off the
current checkout's HEAD (branch repro/<slug>, auto-suffixed on collision),
and runs the agent FROM that worktree so the authored e2e test lands on its
own branch without dirtying your checkout. The worktree is always kept; the
script prints its path + branch + cleanup command at the end.
It lives under dev/ (not shipped in the wheel) rather than as an `omni`
subcommand because it depends on a source checkout — the repro-agent authors
into tests/e2e_ui/ / tests/e2e/, which only exist here.
Also, from PR review:
- AGENTS.md: note that UI-journey reproduction drives the desktop app's
embedded browser, so it expects a desktop / embedded-browser context (fall
back to the backend path / needs_more_info when there's no browser pane).
- README.md: document the dev/repro.py driver.
Co-authored-by: Isaac
* dev/repro-agent: handle compound / multi-symptom bug reports
Ported from the internal repro-agent (omnigent-internal#24). A single bug
report often bundles several distinct symptoms (e.g. "picker unavailable AND
catalog defaults lag"), and they can have different truth on the running
build — one already fixed, the other still live. Averaging them into one
verdict hides the part that's still broken.
AGENTS.md now instructs the agent to:
- enumerate each claimed sub-symptom in Step 1 (don't collapse a compound
report into one journey),
- reproduce and judge each independently in Step 2, and
- roll up to an overall verdict where ANY live sub-symptom ⇒ reproduced
(already_fixed only when every facet is fixed), emitting a per-facet
breakdown (`facets`) in the output so a partial fix stays visible.
Wording adapted to the local variant (running build / local session; no
deployed-app or public-share references).
Co-authored-by: Isaac
* dev/repro: drop the `ref` input — always reproduce against the running build
`ref` never controlled what was validated: the agent always reproduces against
the app it is connected to (the running build / latest main), and `ref` was
only informational — and redundant, since the reported version is already in
the bug report the agent reads. Simplify the input contract to just `bug_url`.
- dev/repro.py: remove the --ref option; the payload is {"bug_url": ...}.
- config.yaml / AGENTS.md / README.md: drop the ref bullet/examples; keep the
guidance that reproduction is always against the running build (so an
old-version report can still land already_fixed).
Co-authored-by: Isaac
A developer-facing repro agent under dev/. Given just a bug (a bug_url — GitHub
issue or Linear ticket — plus an optional ref), it reconstructs the user journey
from the linked report, drives the running Omnigent app it is connected to (the
server `omnigent run` spins up, or one passed with --server) through that journey
until the failure happens live, and authors a durable e2e test (Playwright under
tests/e2e_ui/ for UI bugs, or tests/e2e/ for backend) as the regression artifact.
It reproduces against whatever app it is connected to and authors the test into
the current checkout, so a developer can run it against their own local server:
omnigent run dev/repro-agent -p '{"bug_url":"https://github.com/omnigent-ai/omnigent/issues/1234"}'
It does not fix the bug, merge, or push — it produces a live-confirmed
reproduction plus the test and hands off (the fix half owns the before/after
fail->pass proof).
Files:
- config.yaml — claude-sdk brain, os_env shell/file access, blast-radius guard.
- AGENTS.md — the operating procedure (confirm workspace -> reconstruct journey
-> reproduce live -> author the e2e test -> structured verdict).
- README.md — prerequisites, usage, and what it produces.
Co-authored-by: Isaac
omni host status was slow because _add_daemon_host_status made a
GET /v1/hosts/{id} request for every daemon record, including the many
stale records accumulated over dev sessions (39 in one measured case).
Dead processes can't have an online tunnel, so the correct answer is
host_status=offline with no network round-trip.
Skip the HTTP call when process=offline and set host_status directly.
This cut omni host status from ~14s to ~5s on a workstation with many
stale daemon records.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
When a session's workspace directory no longer exists on the host
(e.g. a worktree was deleted), the host was returning a generic
failed status with no error_code, causing the server to silently
wait out the full connect timeout and then surface a generic
'runner_failed_to_start' banner.
Changes:
- Add WORKSPACE_MISSING_ERROR_CODE ('workspace_missing') to host/frames.py
- Host returns this code when workspace.is_dir() fails, alongside the
existing descriptive error message
- Server (routes_events.py post_event) handles workspace_missing the same
way as harness_not_configured: immediately consumes the user message and
persists an actionable runner_failed_to_start error item with the host's
'workspace path does not exist: ...' message instead of timing out into
a generic RUNNER_UNAVAILABLE
- orchestration.py _ensure_runner_relay_ready skips the connect-timeout
wait for workspace_missing (same as harness_not_configured), and records
the refusal in runner_exit_reports so snapshot-based renders also show
the actionable cause
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(web): stop rendering shell-style env vars in prose as LaTeX math
Error messages like "Unresolved environment variable '$LLM_API_KEY' … Set
$LLM_API_KEY or $OMNIGENT_LLM_API_KEY" render through the assistant markdown
renderer, which has single-dollar math enabled. The paired `$` tokens collapsed
into a garbled inline formula.
normalizeExplicitMathDelimiters already escaped a lone `$` before a digit
(currency); extend that heuristic to also escape shell-style variable
references ($VAR_NAME and ${VAR_NAME}, SCREAMING_CASE) so they stay literal text
instead of flipping the math span.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* docs(web): clarify SHELL_VAR_RE handles single-char braced refs
Address Copilot review: the comment said "2+ chars" but the braced
alternative uses `*`, so `${A}` matches. That's intended — braces
disambiguate a variable reference, so one char is enough there, while the
bare form still requires 2+ so `$X …` reads as inline math. Fix the comment
and add a test for both cases.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): require full-token boundary for bare shell-var match
Address Copilot review: SHELL_VAR_RE's bare branch matched a SCREAMING_CASE
prefix of a mixed-case token (e.g. `$FOOBar$`), escaping the opening `$` while
leaving the closing `$` as a delimiter — an unbalanced span that breaks
genuine inline math. Add a `(?![A-Za-z0-9_])` boundary so only full
SCREAMING_CASE tokens match, and greedy backtracking can't settle on a prefix.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
When the VPN drops, a corporate proxy answers the host tunnel's
WebSocket upgrade with 401/403 before the request reaches the Omnigent
server. `_classify_http_status` treated those as permanently fatal, so a
live, already-registered host exited with code 1 and the user had to
re-run `omnigent host` after reconnecting.
A host that already completed an upgrade proved its credentials and
authorization are valid, so a later 401/403 is almost always a transient
network-path artifact. For a connected host, 401/403 now retries forever
via the normal reconnect path (mirroring the existing login-redirect
design), with a once-per-outage stderr notice so a foreground
`omnigent host` isn't silent. A fresh, never-connected host still fails
loud on the first 401/403.
Fixes OMNI-2367.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
Two fixes to _ManagedMintTokenFactory and _InitialAuthTokenFactory:
1. Only latch declined=True on 400/404 if the factory has never successfully
minted a token. A 400 mid-session (e.g. during an IP ACL flip) is
transient — the server already proved it mints for this runner, so treat
it like any other transient failure instead of bricking the factory.
2. Add a declined property to _InitialAuthTokenFactory that proxies the
inner fallback factory. Without this, auth_flow sees declined=False on
the outer wrapper and raises 'no token' instead of falling back to bare
requests, causing infinite retry loops in PATCH external_session_id and
other callbacks after the inner factory latches declined.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
A managed-host wake (resume_managed_host: resuming a dormant resumable
sandbox on the next message) never forwarded launch-pipeline stages to the
caller, unlike the fresh-launch path (_arm_and_start_host), which threads
on_stage through. As a result _run_managed_wake left the session on the
single "provisioning" band that _kick_managed_wake seeded for the entire
resume — even while the host was already re-execing and dialing back — so
the UI showed a frozen "Provisioning sandbox" band for the whole wake.
Thread on_stage through resume_managed_host into _start_sandbox_host (which
already accepts it), and have _run_managed_wake pass a _publish_sandbox_status
closure. The wake now advances to "starting" (emitted by base start_host)
before "connecting"/"ready", matching a fresh launch.
Signed-off-by: Edwin He <41037314+Edwinhe03@users.noreply.github.com>
Connecting a host to a Databricks-App-deployed omnigent server as a service
principal failed: `omni run --server <app>` resolves credentials through the
Databricks SDK's default chain, which reads only the DEFAULT ~/.databrickscfg
profile. When DEFAULT points at a different workspace than the one fronting the
app, the minted token is for the wrong workspace and the Apps proxy bounces the
request to interactive OIDC (302) instead of admitting it.
Add a `--profile NAME` option to `omni run` that sets DATABRICKS_CONFIG_PROFILE
for the CLI process, so every remote-auth path (_remote_headers, _server_auth,
_DatabricksTokenAuth) resolves the named service-principal profile. This enables
headless M2M access to a deployed app without a prior interactive `omnigent
login`. An explicit --profile wins over an ambient DATABRICKS_CONFIG_PROFILE;
omitting it leaves any preset untouched.
Prereq (Databricks-side, not code): the service principal must have CAN USE on
the app.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(runner): treat HTTP 403 as refreshable on tunnel reconnect
A runner whose auth token expires while the machine is offline can
receive HTTP 403 (not 401) when DNS resolves again and the server
rejects the stale credential. Previously 403 was in
_FATAL_SERVER_HTTP_STATUSES and caused the runner to exit immediately
with no retry, killing any active session.
Move 403 into _REFRESHABLE_HTTP_STATUSES alongside 401. The existing
_handle_refreshable_auth_failure path already handles this correctly:
it attempts one token refresh, and if the factory is invalidatable
(or returns None) the second 403 raises a fatal RuntimeError instead
of looping forever. A runner with no factory still exits fatally on
the first 403.
Add three tests covering the new behaviour:
- 403 with factory → refresh → retry → success
- 403 with invalidatable factory → refresh → persistent 403 → fatal
- 403 without factory → fatal immediately
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): guard 403/401 refresh against transient factory errors
- Drop the inline to_thread(factory) call in the refreshable-status
handler; rely on the loop-top _refresh_auth_token instead, which
already wraps factory calls in try/except for OSError/ValueError.
This prevents a transient IdP error on wake-from-sleep from crashing
serve_tunnel rather than falling back and retrying.
- Also removes the redundant double-refresh-per-cycle that the inline
call introduced.
- Update _handle_refreshable_auth_failure docstring: 401/403 now go
through the streak path, not this function; only 302 redirects
reach it.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(server): import _spawn_archive_stop in routes_core
Missing import introduced in 2ce9c60b.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* perf(sessions): stop and archive in parallel so archiving isn't gated on stop timeouts
Archiving a live session took 5-10s: the sidebar serialized stop -> archive,
and the PATCH handler awaited its own best-effort stop (5s runner / 10s host
teardown ceilings per running session) before flipping the flag — even though
the archive proceeds regardless of the stop's outcome. Fire the client legs
in parallel and detach the server-side stop into a retained background task;
the stop still runs to completion, it just no longer holds the response.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(sessions): let the server own the archive stop so it can't race the client's
Review follow-ups on the parallel-archive change:
- The client no longer sends its own stop_session alongside the archive
PATCH. Two concurrent stops raced the same runner, and because the
runner's stop handlers are not idempotent (kill_session raises once
the pane is gone -> 503), the loser's failure aborted the client stop
before it reached the host-runner teardown -- orphaning a host-spawned
session's dedicated runner. Archive now sends one PATCH.
- The server's detached stop carries the host-runner teardown that only
the client stop used to do, so archiving still drops the runner's
tunnel and flips runner_online. Bulk archive gains this too; it never
sent a client stop.
- The stop is spawned only after the archived flag commits. It ran
ahead of later validations, so a PATCH rejected after that point
(reserved label, runner_id permission) could stop a session it did
not archive.
Adds an e2e_ui browser test for the archive flow plus server coverage
for the teardown and the rejected-PATCH case.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(server): file forked sessions into the source's project
Forking a session filed in a first-class project left the fork unfiled:
fork_conversation built the fork's metadata row without project_id. The
fork route now carries the source's project onto the fork, gated on the
forker owning that project (projects are owner-private, so a fork of a
shared session filed in someone else's project stays unfiled).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): refresh the project folder when a session is forked
A fork inherits the source's project, but the dialog only invalidated the
flat session list — each folder renders its own ["project-sessions", name]
query, which has no poll and converges only on an explicit invalidation.
The push stream can't cover it either: it skips the active session, and the
fork becomes active on navigate. So the clone stayed missing from its folder
until a reload or a re-navigation.
Adds an e2e regression test. It seeds the committed turn the fork action
anchors on straight into the store (new seed_committed_turn helper) instead
of driving a model turn, so it neither waits on nor inherits the flakiness
of the mock-LLM harness.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
web/electron/package.json was deliberately excluded from
update_versions.py because lockstep versions are not valid semver, so
it rotted: v0.7.0 and v0.8.0 shipped a desktop app still calling
itself 0.6.0, and 0.8.1's desktop bump had to be pushed by hand onto
the release branch (and still reads 0.8.0 at the v0.8.1 tag).
Stamp it with the semver translation of the lockstep version instead
(0.6.0rc1 -> 0.6.0-rc.1, 0.7.0.dev0 -> 0.7.0-dev.0, finals
unchanged) — semver orders these the way PEP 440 does (dev < rc <
final), so desktop auto-update comparisons stay correct. check() now
gates the translation, so a drifted desktop version fails the version
lockstep lint. Aligns main's desktop version to 0.9.0-dev.0.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(ci): normalize uv.lock after /regen resolutions
The regen workflow was the one lock-writing CI path left out when the
normalize-then-verify step was added to release/bump/nightly: its
uv lock --upgrade-package runs re-add the size fields the canonical
form forbids, ballooning a /regen'd PR's lockfile diff by ~3k lines of
formatting noise and failing the pre-commit lint on the PR.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(ci): /regen upgrade touches only uv.lock
A targeted Python package upgrade was also deleting and re-resolving
pnpm-lock.yaml from scratch, burying a ~100-line dependency fix under
thousands of lines of npm churn. Plain /regen keeps refreshing both
lockfiles.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* Bump version to 0.9.0.dev0
Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* chore(deps): drop the stale gitpython cooldown exemption
The per-package cutoff (2026-07-24) was added to make 3.1.55 resolvable
while it was inside the P7D window; it aged out, and the frozen cutoff
now excludes 3.1.56/3.1.57, which fix GHSA-p538-c434-8v24 and
GHSA-3f7w-8rr8-f37f — so the OSV audit fails on any PR touching the
lock. The global P7D cooldown admits 3.1.57 on its own now. Lockfile
regen follows via /regen upgrade gitpython.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* chore(oss): regenerate public lockfiles against public PyPI/npm
* chore(oss): regenerate public lockfiles against public PyPI/npm
* chore(deps): normalize the lockfile back to canonical form
The /regen runs regenerate uv.lock without the normalize step the
other lock-writing workflows gained, re-adding the size fields the
canonical form forbids. Text-only cleanup; the resolved versions
(gitpython 3.1.57, aiohttp 3.14.2) are unchanged.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* chore(deps): restore main's pnpm-lock.yaml
The /regen runs regenerate the npm lockfile from scratch even for a
Python-only package upgrade; this PR changes no JS dependency, so
main's lockfile is exactly right for it.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
The auto-generated entry said no user-facing changes: the release's one
change is a cherry-picked revert whose PR is still open against main,
which the changelog curation (merged-PRs-in-range) cannot see.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
The nightly cut is fully unattended, so a broken run blocks nobody and
consumers silently stop getting new builds. Add Nightly Release to the
failure monitor's watch list: its two-consecutive-failures rule and
close-on-green behavior apply unchanged, and skipped quiet nights
conclude success so they close any open tracking issue.
RELEASING.md gains a Nightly builds section: what the workflow does,
how consumers install and update from tags (no PyPI), and that a bad
nightly needs no recovery beyond fixing main.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* feat(harness): add Grok Build (xAI) as a first-class ACP harness
Grok Build (`grok`) had no first-class harness — only usable as a custom `acp:`
agent or as `xai/grok-*` behind openai-agents. Add `harness: grok` (alias
`grok-build`) driving `grok agent stdio` over ACP via the generic AcpExecutor,
the reuse path the issue suggests (like qwen).
- inner/grok_harness.py: thin create_app wrapping AcpExecutor with a fixed
`grok agent stdio` command; auth is Grok's own (grok login / XAI_API_KEY),
Omnigent stores no credential.
- Registry: valid_harnesses / harness_modules / alias grok-build / capabilities
(ACP profile: own-auth, cold resume, SSE permission, interrupt) / label
"Grok Build" / HARNESS_GROK_MODEL.
- Install spec (curl x.ai/cli/install.sh, grok login --device-auth) and
binary-gated readiness, matching the other own-auth CLI harnesses.
Closes#2881
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Signed-off-by: apeltekci <andrew@peltekci.com>
* test(onboarding): include grok spellings in configured-harness-map test
The grok harness added `grok` + `grok-build` to the configured-harness map;
test_configured_harness_map_covers_all_spellings pinned an expected_keys set
that omitted them, so it failed with both as extra items. Add them.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Signed-off-by: apeltekci <andrew@peltekci.com>
* test(e2e): exclude grok from the live no-agent harness matrix
Registering grok as a coding harness added it to the matrix's expected set,
but grok is a headless ACP harness driven over stdio: it authenticates from
the grok CLI's own xAI login rather than the shared gateway/profile probe
wiring, so there is nothing for this binary-less no-agent matrix to probe.
Exclude it alongside goose, which is excluded for the same reason, and name
tests/inner/test_grok_harness.py as its coverage.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Signed-off-by: apeltekci <andrew@peltekci.com>
* fix(harness): drop the grok model-override claim nothing implements
The harness registered HARNESS_GROK_MODEL in model_env_keys, but the executor
never read it, so a spec model or /model pick was silently dropped rather than
applied — and the docstring pointed at a session/set_model path this harness
doesn't implement.
Remove the registry entry and the claim. Grok selects its model in its own CLI;
an Omnigent-driven override for ACP-backed harnesses is a separate concern and
should land with the mechanism that actually applies it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Signed-off-by: apeltekci <andrew@peltekci.com>
* refactor(harness): declarative catalog for builtin ACP CLI harnesses
Promoting an ACP-speaking vendor CLI to a first-class harness has meant
touching 6+ registration points (capabilities, valid set, module map,
aliases, labels, install spec, readiness, setup steps, a per-harness
spawn-env builder, the live e2e matrix exclusion) plus a near-identical
thin inner module. Recent PRs each re-derived this by hand and one shipped
without its spawn-env builder, silently dropping the session cwd and the
spec sandbox.
Add omnigent/acp_cli_harnesses.py: one AcpCliHarness row per vendor CLI
(label, binary, ACP argv, aliases, install and login metadata). Every
registration derives from the row:
- harness_plugins: validity, module routing (all rows run the shared
omnigent/inner/acp_harness.py wrap), aliases, labels, capabilities
(the generic acp profile), install specs and install keys
- onboarding: one-click install allowlist (npm rows) and vendor-login
setup steps derive; readiness rides the existing install-key gate
- runtime/workflow: one shared _build_acp_cli_spawn_env forwarding the
session cwd and serialized os_env, shell-quoting the resolved binary
- runner dispatch: one membership check covers every current and future
row
- tests: readiness spelling lists and the live-matrix exclusion extend
from the catalog; tests/test_acp_cli_harnesses.py drives a fake row
through the builder and dispatch and asserts full registration per
real row
The catalog ships empty; the first rows land with the Grok Build (#3075)
and Qoder (#3560) PRs, each reduced to one dict entry plus docs.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* refactor(grok): ride the ACP CLI harness catalog, one row instead of hand wiring
Rebase the Grok Build harness onto the declarative catalog from
feat/acp-cli-catalog: the thin inner module, the per-registry entries,
the install/readiness edits, and the manual e2e-matrix exclusion all
collapse into one ACP_CLI_HARNESSES row carrying the same label, alias,
command, install hint, and login metadata.
Riding the shared builder also fixes two gaps the hand wiring had: the
session working folder and the spec os_env/sandbox now reach the grok
subprocess (grok_harness.py read HARNESS_GROK_CWD / HARNESS_GROK_OS_ENV
but nothing ever set them), and a resolved binary path containing spaces
survives the shlex-split command string.
Registration, spawn env, readiness gating, setup steps, and the live
matrix exclusion are asserted per row by tests/test_acp_cli_harnesses.py,
replacing tests/inner/test_grok_harness.py.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: apeltekci <andrew@peltekci.com>
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* feat: add nimble_research builtin backed by Nimble Agent API v2
Add a nimble_research built-in tool that delegates a research task to a
Nimble Web Search Agent through the asynchronous Agent API v2: start a
run (POST /v2/agents/{agent_id}/runs), poll it to a terminal status on
a monotonic deadline, then fetch the cited result. The tool returns a
bounded JSON envelope - run id, output (text or structured JSON), and
trust metadata (confidence, sources, per-claim citations) - capped so a
large result cannot blow the model context.
The builtin registers like web_search: a registry factory plus
runner-local dispatch, so a non-OpenAI model's nimble_research call
resolves to the backend. api_key and agent_id come from spec config
(the tool never creates agents; one-time bootstrap is documented in the
module); errors are returned as strings and always carry the run id,
including timeout, failure, cancellation, and unknown-status paths.
Polling honors Retry-After on 429 and retries transient failures within
a bounded budget; run creation is never retried.
Includes unit, dispatch, and e2e tests (respx transport mocks and a
fake-clock seam; the e2e drives the full lifecycle against a local
Agent API stub).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* feat: add nimble_extract builtin backed by Nimble Extract Templates
Add a nimble_extract built-in tool that runs one of the account's Nimble
extract templates (POST /v2/extract/templates/run) and returns the
template's structured, parsed results as JSON in one synchronous call.
The template is named in spec config; the LLM supplies the template's
params (each template declares its own input schema, discoverable via
GET /v2/extract/templates/{name}).
This is the migration target for the deprecated one-call /v1/agent
site-scraping path: same structured-entities output contract, now on
the current Extract Templates API. The predecessor tool name is retired
rather than aliased - the registry does not reserve it, and a test
locks that in - so the old name can never silently point at a
different API.
Wiring mirrors nimble_research: registry factory plus runner-local
dispatch. api_key and template come from spec config; errors are
returned as strings with the template named and the server's task id
preserved for supportability (parsing failures, template-not-found,
params rejection, and server error bodies are each mapped to clear
messages); output is capped to keep the model context bounded.
Includes unit, dispatch, and e2e tests (respx transport mocks; the e2e
drives the flow against a local Extract Templates stub), with
captured-request assertions that every request carries the
X-Client-Source header.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* fix(nimble_research): harden malformed-config and envelope bounds
Catch httpx.InvalidURL when building the run URL so a control character
in the configured agent_id returns the builtin's own clean error string
instead of escaping its documented never-raises contract (agent_id is
interpolated into the run URL path).
Cap each API-supplied trust string - reasoning, source and citation url
and title, and the output type - so a single oversized value cannot
inflate the returned envelope past its intended bound, matching the
list-length caps already applied to sources, claims, and citations.
Adds tests: a control-char agent_id returns an error with no request
made, and an oversized trust.reasoning is capped with the envelope
still valid JSON.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* fix(nimble_research): complete the never-raises and envelope bounds
Follow-up to the previous hardening pass, which covered only part of each
surface.
Never-raises: httpx.InvalidURL is not a subclass of RequestError, so it
also had to be handled on the poll and result hops. The run id comes from
the API and is only prefix-validated, so a control character after the
prefix could raise out of the tool. Polling treats it as permanent and
returns immediately rather than spending its transient-retry budget on an
error that cannot become valid.
Envelope bounds: cap the remaining API-supplied strings that reached the
envelope uncapped - trust confidence, per-claim path and confidence - and
drop a non-string source or citation url instead of passing the raw value
through. Also cap API-supplied text reflected into error strings, which
could otherwise be arbitrarily long.
Adds regression tests for both hops, for every capped field, for the
dropped non-string url, and for an oversized server error message.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* fix(nimble_research): bound run status, run id, and the trust section
The result body's run status was accepted as any string and interpolated
raw into the failure message, so a malformed status could turn into a
multi-megabyte error string. Only a known terminal status is trusted now,
and the message caps the values it reflects.
Bound the accepted run id at creation instead of echoing an arbitrary one
through later messages, and cap the trust section as a whole: the
per-field caps still multiplied across sources, claims and citations.
Includes regression tests for each bound.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* feat(nimble_research): adopt nimble-python 1.2 typed run fields
Create runs through the released nimble-python 1.2.0 client instead of
hand-rolled requests, and expose the typed per-run fields it added.
agent_id is now optional and selects the create route: when it is omitted
the run is created with agents.run() so Nimble provisions the agent, and
when it is set the run is created with agents.runs.create() against that
agent. Both routes forward input_data, output_schema, sources, agent_name,
skill, and use_case as typed arguments, so no extra_body escape hatch is
needed. The client is built with max_retries=0, because creating a run is
billable and not idempotent and the API exposes no idempotency key.
effort stays an optional override, so leaving it unset lets the selected
agent or template default apply. low, medium, high, and x-high are
selectable per run. max is a coming-soon custom-budget tier: it stops with
a pointer to the Nimble product team, and only degrades to x-high when a
spec opts in explicitly. The degradation is reported on every outcome, so
a run that was downgraded and then failed still says so.
The agent id returned by creation addresses the rest of the lifecycle,
since on the generated route it is the only one that exists, and a run
that comes back owned by a different agent is rejected rather than
retargeted. Identifiers are checked against an allowlist before they are
interpolated into a request path. Status polling defaults to ten seconds.
Includes unit tests.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* fix(nimble_research): warn against resubmitting an unresolved create
A create that fails in transport or times out may still have been received,
and a 408 or 5xx reached Nimble before the failure was reported, so the run
can be live and billed while the call reports an error. A 202 carrying an
unusable body is the settled version of the same problem: the run exists,
but the response cannot address it.
All of these now say so and tell the caller not to resubmit, since a
resubmission pays for the task a second time. The guidance names the run id
when one survived, and points at the account's recent run history when none
did. A clear rejection still carries no such warning: 401, 403, 404 and 422
create nothing, and attaching the warning to them would only teach the
reader to skip it.
Includes unit tests for the ambiguous and settled paths, and for the
rejections that must stay silent.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* fix(deps): upgrade GitPython to 3.1.55 to clear advisories
The lockfile pinned GitPython 3.1.50, which carries eight advisories whose
fixes land across 3.1.51, 3.1.53, 3.1.54 and 3.1.55. The dependency audit
only runs when the lockfile changes, so the pin was invisible until it was
touched, and then it failed the scan.
3.1.55 is the first release that clears all eight. It sits one day past the
P7D resolution window, so it needs a per-package exception alongside the
existing ones; the cutoff is set to land on 3.1.55 rather than the latest
release, keeping the change to the smallest version that resolves the
advisories.
GitPython is a transitive dependency, so this is a lockfile-only change and
no declared requirement moves. No other package version changes.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* fix: align Nimble 1.2 run controls with released contract
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* fix(nimble_research): never invite a resubmit of a billed run
Once run creation succeeds the run exists and has been billed, but only the
create path said so. The 409 branch told the caller to "retry the task to
fetch it" — on a run already observed complete — which reads as an instruction
to call the tool again and pay for a second run to read the first one's
result. Timeout, polling and result-fetch failures said nothing either.
Every post-create failure now ends with the same guidance the create path
gives, keyed to the run id: do not resubmit, reconcile the run that already
exists. A create-time 429 stays a clear rejection, since a rate limiter
refuses the request before a run is started; that classification is now
documented and covered.
Also drops the notice channel left behind when the effort downgrade was
removed. _resolve_effort returned None for it at every exit, so the value was
always None and the code that consumed it was unreachable; a resolved effort
is now simply what the caller asked for. The tool schema's sources object is
tightened to match what the tool already enforces, so a schema-conformant call
is not rejected at runtime.
Includes unit tests for each post-create path and the rejection that must stay
silent.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
* feat(onboarding): advertise the nimble builtins to the agent builder
list_builtin_tools.py is the onboarding assistant's sole source of truth
for recommendable builtins; without these entries the assistant can
never surface nimble_extract or nimble_research when building an agent.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* refactor(deps): move nimble-python behind a `nimble` extra
nimble-python was a baseline dependency, so every install pulled a
partner SDK that only the nimble_research builtin uses (nimble_extract
talks raw httpx). Follow the hindsight-client pattern: the SDK moves to
an optional `nimble` extra, nimble_research imports it lazily inside
_start_run and reports which extra to install (checked before anything
is sent, so nothing is billed), and the onboarding catalog advertises
the tool only when the SDK is importable. The client stays in the dev
set so the credential-free suites keep driving the real SDK, and mypy
gets the same ignore_missing_imports override as the other lazy-import
extras.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(nimble): address Polly review findings
Blocking items, all verified before fixing:
- Guard use_case with isinstance before the frozenset membership test; a
list/dict argument raised TypeError (unhashable) out of invoke(),
breaking the never-raises contract. Now a clear tool error, unbilled.
- Catch APIError (e.g. APIResponseValidationError, which subclasses
APIError, not APIStatusError/APIConnectionError) in the create path
and route it through the unresolved-create guidance: a 2xx whose body
fails SDK validation means the run may exist and be billed, which is
exactly the case the do-not-resubmit warning exists for.
- Clamp each HTTP call's timeout to the remaining deadline via
_request_timeout, so a single create/poll/result request can no longer
overrun the tool's documented timeout_seconds budget.
Also apply the research module's error-string caps to nimble_extract
(message, task id, parsing detail, status), closing the one reflected
uncapped path Polly's non-blocking notes and the maintainer review both
flagged. Regression tests for all four.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: Kobi Kadosh <kobi.kadosh@gmail.com>
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Dhruv Gupta <dhruv.gupta@databricks.com>
The release cut, main bump, and nightly cut all regenerate uv.lock in
CI. The runner's uv now writes size fields on file entries, which the
repo's canonical lockfile form (scripts/normalize_uv_lock_registry.py,
enforced by the pre-commit hook) forbids — so the v0.8.0 release
commit went red on the branch-push lint run, and the next cut from
that branch would fail the green-CI gate. Run the normalizer after
uv lock (fixer exits non-zero when it rewrites, so tolerate that),
then hard-verify with --check so a genuinely broken lockfile still
fails the step.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Promoting an ACP-speaking vendor CLI to a first-class harness has meant
touching 6+ registration points (capabilities, valid set, module map,
aliases, labels, install spec, readiness, setup steps, a per-harness
spawn-env builder, the live e2e matrix exclusion) plus a near-identical
thin inner module. Recent PRs each re-derived this by hand and one shipped
without its spawn-env builder, silently dropping the session cwd and the
spec sandbox.
Add omnigent/acp_cli_harnesses.py: one AcpCliHarness row per vendor CLI
(label, binary, ACP argv, aliases, install and login metadata). Every
registration derives from the row:
- harness_plugins: validity, module routing (all rows run the shared
omnigent/inner/acp_harness.py wrap), aliases, labels, capabilities
(the generic acp profile), install specs and install keys
- onboarding: one-click install allowlist (npm rows) and vendor-login
setup steps derive; readiness rides the existing install-key gate
- runtime/workflow: one shared _build_acp_cli_spawn_env forwarding the
session cwd and serialized os_env, shell-quoting the resolved binary
- runner dispatch: one membership check covers every current and future
row
- tests: readiness spelling lists and the live-matrix exclusion extend
from the catalog; tests/test_acp_cli_harnesses.py drives a fake row
through the builder and dispatch and asserts full registration per
real row
The catalog ships empty; the first rows land with the Grok Build (#3075)
and Qoder (#3560) PRs, each reduced to one dict entry plus docs.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* feat(ci): nightly prerelease builds — tag-only dev-datestamp cuts from main
Adds nightly-release.yml: every night at 04:30 UTC it walks main to the
newest commit with completed green CI, stamps the lockstep version to
X.Y.Z.devYYYYMMDD, commits the stamp detached on top of that base, and
pushes only the tag via the omnigent-ci App token. Quiet nights (no new
commits since the last nightly tag) and same-day reruns no-op.
Deliberately not the release.yml flow: no release branch, no main bump,
no benchmark gate. Downstream is already dev-quiet: no GitHub release,
no notes, no changelog, no homebrew; images publish the immutable
version tag; update-check ignores dev releases and omni upgrade --pre
opts in. The datestamp is fixed-width because PEP 440 compares the dev
segment as one integer — a wider stamp would sort above every narrower
one forever.
PyPI publishing follows separately via the secure release repo's
scheduled lane.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* feat(ci): nightly consumer update script; drop the stale PyPI hand-off note
scripts/update_nightly.sh resolves the newest vX.Y.Z.devYYYYMMDD tag
(version sorts before date, so the first nightly after a main version
bump outranks all older ones; the 8-digit date requirement screens out
legacy .dev0-style tags) and installs it with uv, pinning the lockstep
trio to that one tagged commit. Idempotent, so it is cron-safe: it
exits fast when the newest nightly is already installed instead of
redoing the web-UI build.
The workflow header no longer claims the secure release repo publishes
nightly tags to PyPI: that lane was dropped, nightlies are consumed
straight from the tag.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* feat(cli): omni upgrade --nightly moves onto the newest nightly tag
Nightlies are vX.Y.Z.devYYYYMMDD git tags that never reach the package
index, so the flag answers 'is there something newer' from the repo's
tags (git ls-remote + PEP 440 max, so the first nightly after a main
version bump outranks all older ones) and reinstalls with a git spec
pinned to that tag, per installer (uv/pipx/pip/poetry). It dispatches
before the VCS-vs-registry split: a registry install hops onto the
channel, and a VCS install pinned to an older nightly moves tags
instead of re-pulling its pinned ref. Same drain/stop, --check, and
probe-the-disk verification contracts as the release path.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
## Related issue
N/A
## Summary
- After the switch to corepack/pnpm, `pip install .` / `uv sync` could hang
indefinitely for users who have a corepack `pnpm` shim on PATH but have
never downloaded pnpm. Corepack prints `! Corepack is about to download
.../pnpm-11.15.1.tgz` and then blocks on `? Do you want to continue? [Y/n]`.
Build backends capture output, so the prompt is invisible and the install
just sits there until the 600s timeout.
- The trigger is the shim, not the `corepack pnpm` fallback: corepack's
`dist/pnpm.js` does `COREPACK_ENABLE_DOWNLOAD_PROMPT ??= '1'` while explicit
`dist/corepack.js` uses `'0'`. `shutil.which("pnpm")` finds the shim, so the
prompting path is the one that looked fine. CI is unaffected because corepack
skips the prompt when `$CI` is set.
- Run both pnpm commands in `setup.py` with
`COREPACK_ENABLE_DOWNLOAD_PROMPT=0` (download without asking) and
`stdin=DEVNULL` so nothing else in the toolchain can block on input we can
never deliver. Applied the same fix to `tests/e2e_ui/conftest.py`, which had
the identical latent hang under captured pytest output.
## Test Plan
Reproduced the hang and verified the fix against the pinned `pnpm@11.15.1`,
handing the child a real TTY via `pty.openpty()` and an empty `COREPACK_HOME`:
```
BEFORE (shim default prompt=1, TTY stdin): HUNG (timeout)
err='! Corepack is about to download .../pnpm-11.15.1.tgz\n? Do yo'
AFTER (prompt=0 + stdin=DEVNULL): proceeds straight to download
```
End-to-end check of the install path:
```bash
rm -rf ~/.cache/node/corepack "$COREPACK_HOME"
corepack enable # pnpm shim on PATH, pnpm not yet fetched
rm -rf omnigent/server/static/web-ui
pip install . # previously stalled with no output
```
`ruff check` / `ruff format --check` clean on both files.
## Demo
N/A
## Type of change
- [x] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Verified manually: the failure only reproduces with a corepack `pnpm` shim, an
unpopulated `COREPACK_HOME`, and a TTY on stdin, so an automated test would
have to stand up a pty plus a registry fetch inside the build backend. Covered
instead by the pty-based before/after check in the Test Plan.
## Changelog
`pip install` / `uv sync` no longer hangs when pnpm is provided by a corepack
shim that has not downloaded pnpm yet.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
When omnigent-slack joined the lockstep, the cut job's hand-kept git
add list kept staging only the original five paths, so the release
commit shipped integrations/slack/pyproject.toml unstamped. At the
v0.7.0 tag the tree pins omnigent-slack==0.7.0 while the in-tree
package still says 0.7.0.dev0: uv sync --locked fails at the tag, a
source install with the slack extra cannot resolve, and lint went red
on both release/v0.7.0 pushes without blocking the tag. Stage with
git add -A like bump-version.yml so the staged set tracks whatever
update_versions.py stamps.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* Add WebSocket load test (dev/loadtest/) + run-load-test skill
Adds a Locust load test that opens N concurrent WebSocket connections to
WS /v1/sessions/updates and holds them open, measuring the server's
WebSocket fan-out (handshake, origin/auth gating, watch-set diffing,
heartbeat) under concurrency — no runner, LLM, or agent turns.
- dev/loadtest/ws_load_test.py: the locustfile (SessionUpdatesUser).
- dev/loadtest/run.py: runner taking server + host + load params, runs
locust headless, and writes a result set (summary.md, CSV, HTML, config).
- loadtest extra (locust + websocket-client) in pyproject.toml + uv.lock.
- .claude/skills/run-load-test: skill that gathers inputs, runs, and
explains the latency results.
Co-authored-by: Isaac
Signed-off-by: Shivam Mittal <shivam.mittal@databricks.com>
* Launch locust via sys.executable -m locust in the load-test runner
run.py launched locust as a bare `locust` command, which resolves through
PATH and can pick up a stale/broken locust from a different Python (e.g. a
~/.local 3.10 install missing gevent's zope.event) even when run.py itself
runs under a venv — crashing the run with ModuleNotFoundError before locust
starts. Launch it as `sys.executable -m locust` so it always uses the same
interpreter + site-packages that run.py runs under. Preflight now checks
importlib.util.find_spec (the actual interpreter) instead of shutil.which
(PATH), and --web execs sys.executable too.
Co-authored-by: Isaac
Signed-off-by: Shivam Mittal <shivam.mittal@databricks.com>
* Genericize --mount-prefix docs to reverse-proxy sub-paths
Replace deployment-specific mount-prefix details with a provider-neutral
"behind a reverse proxy at a sub-path" framing (neutral /omnigent example)
across the README, the run-load-test skill, and the run.py / ws_load_test.py
help + docstrings. The --mount-prefix flag itself is unchanged.
Co-authored-by: Isaac
Signed-off-by: Shivam Mittal <shivam.mittal@databricks.com>
* Add runner-level turn load test (real multi-turn conversations, mocked LLM)
turn_load.py drives real agent turns through the runner — the full
POST .../events → server → runner → executor → LLM → stream → idle loop —
under concurrency, with the LLM mocked (zero latency) so the numbers isolate
Omnigent's own per-turn / history-handling overhead. Runs N concurrent
conversations of M sequential turns each on one durable session, so history
grows across the turns (a real long conversation, not N one-shots).
It boots the whole stack itself (server + zero-latency mock LLM + runner) by
reusing the benchmark harness's BenchEnvironment, using the in-process
openai-agents harness — no vendor CLI, no real API key — so it runs from a repo
checkout with no server to point at. Concurrency is asyncio (the runner stack
is async), not Locust. Writes the same summary.md / run_config.json result
format as the WS runner.
Documents both scenarios (WebSocket fan-out vs runner turns) in the README and
the run-load-test skill.
Co-authored-by: Isaac
Signed-off-by: Shivam Mittal <shivam.mittal@databricks.com>
* Address review: fix socket leak, URL/timeout edge cases, double-count; add tests
Copilot review follow-ups on the load-test harness:
- ws_load_test: assign self.ws before the send/recv steps so a post-create
failure closes the socket instead of leaking it.
- ws_load_test: _ws_url now treats a schemeless host (localhost:8000, which
Locust accepts) as ws:// rather than emitting an invalid URL.
- ws_load_test: _read_until_snapshot caps each recv to the remaining deadline
so a late frame can't overrun by a full read timeout.
- ws_load_test: WS_READ_TIMEOUT falls back to the default on a non-numeric
value instead of raising in on_start.
- run.py: preflight websocket-client as well as locust; rename _fmt_ms ->
_fmt_num (it also formats Requests/s).
- run.py: _write_summary skips locust's Aggregated row when totaling, which was
double-counting the headline request/failure counts.
- docs: the scenario reads AUTH_TOKEN from the environment; drop the wrong
`-e AUTH_TOKEN` locust-flag examples (AUTH_TOKEN=... locust ...).
- tests: add tests/loadtest unit tests for the pure helpers (URL/env/argv
wiring, summary formatting, timeout parsing) — deterministic, no server boot.
Co-authored-by: Isaac
Signed-off-by: Shivam Mittal <shivam.mittal@databricks.com>
* Redesign as one load test: each user is a real host driving real turns
Collapse the two scenarios (ws_load_test.py + turn_load.py) into a single
load test where each Locust user IS a real omnigent host. Each user spawns a
real `omnigent host` subprocess (unique identity + per-host $HOME so the
host-daemon singleton guard doesn't collide), registers it over the host
tunnel, then creates host-bound sessions and drives real multi-turn
conversations — every turn is a genuine post→idle loop through a runner the
host spawns, with the LLM mocked (zero latency). `-u N` scales the number of
hosts; Locust does the concurrency.
run.py boots the whole stack (server + mock LLM via BenchEnvironment),
registers one agent, sets the mock reply, then runs Locust against it — there
is no --server to pass, since mocking the LLM requires a stack we control. It
reuses the CSV→summary.md machinery (Aggregated-row dedupe kept).
Capacity-limited by design: N hosts × M sessions = N×M real runner processes on
the load box, so it drives genuine end-to-end turns rather than faking the
runner, but does not scale to hundreds on one machine (documented). Removes the
websocket-client dep (no longer used); needs [loadtest,dev,agents-sdk]. README,
skill, and tests updated for the single scenario.
Verified locally: 5 hosts × 3 turns → 133 turns, 0 failures.
Co-authored-by: Isaac
Signed-off-by: Shivam Mittal <shivam.mittal@databricks.com>
---------
Signed-off-by: Shivam Mittal <shivam.mittal@databricks.com>
Co-authored-by: Shivam Mittal <shivam.mittal@databricks.com>
PR #3420 validated the OpenClaw Gateway ACP path end-to-end against a live
Gateway, but docs/openclaw.md still read as if streaming/final replies were
only protocol-matched and the integration provisional. Update the
compatibility section to state what live validation confirmed — streaming
assistant replies, native tool execution, ACP permission routing, and session
resume — and reframe the remaining Control-UI-sync gap as a known limitation
rather than an open question. Keep the note that CI cannot run OpenClaw.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* fix(ap-web): harden math rendering
Load KaTeX runtime styles in every web entrypoint and normalize common TeX delimiters so streamed formulas, radicals, and display math render reliably across chat surfaces.
Co-authored-by: Cursor <cursoragent@cursor.com>
Signed-off-by: buyicoder <169354621+buyicoder@users.noreply.github.com>
* fix(ap-web): make math delimiter normalization region-aware
Address Polly review notes on the math-rendering hardening:
- Skip normalization inside existing $…$/$$…$$ spans and treat a
literal backslash-backslash as a verbatim escape, so a LaTeX line break
like \\[1em] inside an aligned display block is no longer mistaken for
a \[ opener and corrupted.
- Track backtick-run length so \(/\[ inside a multi-backtick inline-code
span is left verbatim.
- Correct the stale FILE_PATH_AWARE_COMPONENTS comment now that the memo
comparator is gone and MessageResponse shallow-compares props.
Co-authored-by: Isaac
* fix(ap-web): guard currency dollars and indented fences in math normalizer
Follow-up on Polly review notes:
- A single $ immediately before a digit reads as currency ($5), so it is
escaped and does not flip the math-span toggle. Prevents prose like
"it costs $5 or $10" from parsing as inline math now that
single-dollar math is enabled globally. An escaped \$ is copied verbatim.
- Fence detection now allows CommonMark's 0-3 leading spaces and matches the
full fence run, so an indented ```-fenced block containing \(...\) is not
normalized (and a 4-backtick run no longer leaks into inline-code tracking).
Co-authored-by: Isaac
* fix(ap-web): use String.match for fence detection to clear exfil scan
The security Exfil scan flags RegExp.prototype.exec() because its text-only
regex matches the substring 'exec(', which is meant to catch Python dynamic
code execution (exec/eval/__import__). This is a pure in-memory regex match
against local string data, so switch to the equivalent String.match(), which
returns the same match array for a non-global regex and avoids the token.
Co-authored-by: Isaac
* fix(ap-web): address Copilot review on math normalizer and styles
- Track the opening fence marker so a fenced code block closes only on a
matching fence char with a run at least as long (CommonMark). A stray
`~~~` line inside a ```-fenced block no longer flips the fence off and
lets math normalization run inside code.
- Drop the no-op `overflow-y: visible` on `.katex-display`; with a
non-visible overflow-x the browser computes overflow-y as auto anyway, so
it only risked stray vertical scrollbars.
- Resolve the entrypoint-style guard test's paths from import.meta.url
instead of process.cwd() so it doesn't depend on the runner's directory.
Co-authored-by: Isaac
---------
Signed-off-by: buyicoder <169354621+buyicoder@users.noreply.github.com>
Co-authored-by: zhanyongjie <zhanyongjie@gmail.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Serena Ruan <serena.rxy@gmail.com>
* fix(pi): surface credential resolution error when gateway provider's env var is unset
When a `kind: gateway` provider is configured as the pi harness default
via `default: pi` and its `api_key_ref: env:VAR` cannot resolve (because
VAR is not exported in the runner's environment), `_optional_provider_family`
previously caught the OmnigentError from `resolve_secret` and returned None
silently. The outer `_apply_provider_to_pi` then raised a generic
"no family whose credentials resolve — set the api_key env var for its
'anthropic' or 'openai' family" message with no mention of which specific
variable to export, making the error hard to act on.
Change `_optional_provider_family` to return the captured error alongside
None (as a tuple), and surface that error in the "no family resolves"
message so the user sees exactly which env var (e.g. `$MY_TOKEN` from
`api_key_ref: env:MY_TOKEN`) needs to be set.
The design intent of the silent catch is preserved: a family whose key is
unset is still treated as absent so pi can fall back to the other family
when only one key is exported. The only change is that the fallback-failure
error now carries the root cause.
Closes#3788
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* address review: fix return type annotation, correct keychain docstring, remove issue refs from tests
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): forward provider api_key_ref env vars into runner subprocess
_build_runner_env filters the host environment before spawning the runner
subprocess, passing only an allowlist of known credential vars
(ANTHROPIC_API_KEY, OPENAI_API_KEY, etc.). A user who configures a gateway
provider with a custom env var via api_key_ref: env:MY_TOKEN would find that
MY_TOKEN is present in their shell and daemon process but stripped before
reaching the runner — resolve_secret then fails, _optional_provider_family
returns None for the family, and _apply_provider_to_pi raises the no-family-
resolves error.
Add provider_credential_env_vars(config) to provider_config.py, which scans
all inline-family providers for api_key_ref: env:VAR and api_key: $VAR
references and returns the set of env var names (plus OMNIGENT_-prefixed
aliases). Wire this into _build_runner_env so those vars are automatically
forwarded alongside the standard HARNESS_CREDENTIAL_ENV_VARS, without
requiring users to list them in OMNIGENT_RUNNER_ENV_PASSTHROUGH by hand.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): add authHeader to generic openai provider entries in models.json
Generic (non-Databricks) OpenAI-compatible gateways expect
Authorization: Bearer <token>. The 'databricks' and 'databricks-completions'
provider entries in the generated models.json were missing authHeader: True
on the generic provider path, so Pi used the Databricks-native auth scheme
instead — causing a 401 Missing Authentication header from the gateway.
Add authHeader: True to both entries when is_generic_provider is true,
matching the pattern already used by databricks-openai, databricks-anthropic,
and databricks-mlflow.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi-native): qualify namespaced model ids in --model arg to prevent builtin routing
When a gateway provider's model id contains a slash (e.g. an OpenRouter
namespaced id like 'moonshotai/kimi-k2.5'), Pi's arg parser treats
'provider/model' in --model as a provider override, routing to the builtin
'moonshotai' provider instead of our custom 'omnigent' provider. The builtin
has no API key, producing 'No API key for provider: openai-codex'.
Pass the fully-qualified 'provider/model' form (e.g.
'omnigent/moonshotai/kimi-k2.5') when the model id contains a slash, so
Pi's findExactModelReferenceMatch matches the canonical form under our
provider first.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(worktree-guard): switch to posixpath for path normalization to ensure consistent behavior across platforms
* Windows-only escape in worktree_guard, the sole write confinement for unsandboxed workers: it reasoned in POSIX but normalized with os.path, which is ntpath on Windows and rewrites / to \ — so startswith("/") never fired and /etc/passwd returned ALLOW.
Fixed by normalizing with posixpath explicitly, plus a drive-letter reject for C:/Windows/x, which posixpath reads as an ordinary relative dir named C:.
Two follow-ups from Copilot: the drive check ran on the raw path, so ./C:/… (and a/../C:/…) normalized past it — moved it after normalization; and isalpha() narrowed to ASCII, since Windows drives are [A-Za-z] and the Unicode form over-rejected.
109 passed on Windows, where four of those cases fail on main. Audited environment_filesystem.py:190 in the same pass — it pairs normpath with os.path.isabs, which holds on both platforms, so it needs no change.
* feat(web): move Chat/Terminal switcher into the header
Terminal-first sessions previously toggled between chat and terminal via
an in-page pill above the composer. Replace it with a MessagesSquare +
chevron icon button in the ChatHeader (next to the agent-info icon) that
opens a Chat/Terminal dropdown, freeing the composer area and keeping the
switcher with the other session controls.
The new ViewModeToggle reads the same TerminalFirstContext the pill did,
so behavior is unchanged: it self-gates for non-terminal-first sessions,
the iOS shell (native Liquid Glass bar), and rail-opened shell views, and
disables the Terminal option (with a spinner while starting up) until a
PTY is reachable. A tooltip names the current view. Removes the pill, its
dead CSS, and the now-redundant iOS keyboard guard.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): update e2e locators + a11y for header view toggle
The render-parity e2e helpers located the old in-page pill via
`role="group" name="View mode"` and clicked its inner Chat/Terminal
buttons. The header switcher is a dropdown, so point them at the
`view-mode-toggle` trigger and click the Chat/Terminal menuitemradio.
Also address review feedback on ViewModeToggle: import the shared
`TerminalFirstView` type instead of a duplicated union in the setView
cast, and only suppress dropdown close-refocus for pointer closes so
keyboard/AT users keep their place (mouse closes still avoid the stuck
ghost-button focus ring).
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(codex-native): tear down app-server when TUI pane is reaped or exits
Each codex-native session (Polly dispatches every codex sub-agent this
way) runs two codex processes on the runner: the codex app-server backend
and the codex --remote TUI pane. Only DELETE /v1/sessions ran the full
cleanup that cancels the forwarder and closes the app-server. Two other
ways the TUI pane goes away left the app-server orphaned for the runner's
lifetime:
- the idle pane reaper closes the tmux pane after the idle window but
never touched _AUTO_CODEX_APP_SERVERS, and
- an unexpected TUI exit (crash / OOM / host recycle) evicted the pane
without cancelling the forwarder.
On a long-lived multi-session runner, every idle or crashed codex
sub-agent leaked a codex app-server process — the pile-up reported in
omnigents-qa.
Add teardown_codex_native_app_server(session_id): cancel the session's
forwarder (whose finally closes the app-server) and close any leftover
registered server. It's a no-op for a session with no registered codex
app-server, so it's safe to call from the shared pane-teardown paths for
every harness. Wire it into the reaper's reap and the terminal-exit
publisher.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(codex-native): reap codex app-server even if pane close raises
Move the codex app-server teardown in the idle-pane reaper into the
finally block. close_terminal() can propagate (TerminalInstance.close()
raises anything but TimeoutError), and in that partial-failure mode the
teardown line in the try body was skipped — leaving the exact orphaned
app-server this fix targets. The helper is idempotent and suppresses its
own errors, so running it in finally never masks the original exception.
Addresses Copilot review on #3925.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(codex-native): close app-servers on host/runner stop + boot reconcile
Host-spawned codex app-servers are spawned start_new_session=True, so
they survive their runner's death. Two gaps left them orphaned:
- On a graceful host/runner stop the host SIGTERMs the runner without a
per-session DELETE /v1/sessions, so per-session teardown never fired and
_stop_pm never closed _AUTO_CODEX_APP_SERVERS — every host-spawned codex
app-server leaked even on a clean stop. (The TUI panes were already
closed by the terminal registry's shutdown; only the app-server half
leaked.)
- On a hard death (SIGKILL / OOM / crash) nothing runs at all, and the
crash-safe registry was only reconciled when a NEW codex session
started — so orphans lingered until the next codex launch, if ever.
Add teardown_all_codex_native_app_servers() and call it from _stop_pm so a
graceful stop takes the app-servers down with the runner. Add a boot-time
reconcile_codex_native_process_registry() in _start_pm so a fresh runner
reaps orphans a dead predecessor left (owner-lock held => live sibling,
skipped). Reconcile runs in a thread since it does blocking file/PID work.
The --remote TUI self-exits when its app-server dies (observed: every
orphan seen in the field was an app-server, zero orphaned TUIs), and the
graceful path already closes TUI panes, so no tmux-name plumbing is added.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(cli): add `omnigent diagnose` environment snapshot
Add a read-only `omnigent diagnose` command that prints a small, secret-free
environment snapshot for bug reports: CLI version, OS/Python, and the server's
auth mode. With `--server <url>` (or a resolvable configured/local server) it
reads the server version and real auth mode from the unauthed `GET /v1/info`
endpoint, so version skew between CLI and server is visible — the same reason
the session-info popover shows `server · host`.
The auth mode doubles as the OSS-vs-managed signal: accounts | single_user |
oidc | header, derived from `/v1/info` when the server is reachable and falling
back to the local environment otherwise (tagged by `auth_source_origin` so the
two are never confused). The snapshot carries no secrets — only versions, OS,
and the coarse auth mode.
`omnigent doctor` (install-ledger migration) is left untouched.
Co-authored-by: Isaac
* fix(cli): address diagnose review — redact server_url, e2e test, help caution
Review follow-ups on the `omnigent diagnose` PR:
- Redact userinfo and query/fragment from the reported `server_url` so a
`--server https://user:pass@host` value can't leak credentials into the
snapshot (the "safe to paste into an issue" invariant).
- Add CLI-level tests (CliRunner + respx over /v1/info) exercising the command
wiring and output format end-to-end, alongside the existing unit tests.
- Note in `--help` that `--server` should point only at a trusted server, since
reaching a managed server may attach stored/ambient credentials to the request
(same behavior as `session export` / `run --server`).
Auth is intentionally still attached to the /v1/info probe: a managed server
sits behind an auth proxy that 401s an unauthenticated request, so dropping it
would break the OSS-vs-managed signal for exactly the managed case. Attaching
credentials to the request does not put secrets in the output, which is what the
"secret-free" guarantee covers.
Co-authored-by: Isaac
* fix(cli): harden diagnose URL redaction + register in subcommand allowlist
- _redact_url: fix two leaks the review found. Scheme-less inputs with userinfo
(`user:pass@host:6767`) were returned unchanged because urlsplit reads the
`user:` as a scheme — now scrubbed. IPv6 literals lost their required `[...]`
brackets when netloc was rebuilt from hostname/port — now the userinfo is
dropped off the authority in place, preserving brackets and host casing.
- Add `diagnose` to `_CLICK_SUBCOMMANDS` so `omnigent diagnose` is reachable
from main() (a registered command missing from the allowlist is rejected as
removed ad-hoc chat). Fixes test_click_subcommands_allowlist_covers_registered_commands.
Co-authored-by: Isaac
* fix(cli): make diagnose URL redaction leak-proof on malformed/scheme-less input
Follow-up on review: _redact_url used urlsplit, which raises ValueError on a
malformed IPv6 URL (the fallback then returned the raw string, leaking any
user:pass@) and left query/fragment intact on scheme-less inputs. Rewrote it as
pure string surgery — cut at the first ?/#, then drop a user:pass@ prefix from
the authority — so credentials and tokens are stripped uniformly regardless of
URL shape, with no parser that can raise. IPv6 brackets and host casing are
preserved.
Co-authored-by: Isaac
The Projects group-header kebab (⋯) rendered next to "New project" even
when its menu had no items to show. With no projects filed, neither the
expand/collapse controls (need projectNames.length > 0) nor "Select
sessions" (needs project sessions) apply, so the menu opened empty.
Gate the kebab on whether either item is available, leaving only the
"New project" button when there's nothing to offer.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
The "My sessions" / "Shared with me" tabs were hidden whenever bulk
selection mode was active, stranding the viewer on whichever scope they
happened to be on. Keep the tabs visible during selection so the scope
stays switchable.
Selection is a single global set while the tabs show disjoint,
ownership-scoped slices, so changing the visible tab now exits selection
mode — otherwise the bulk-action bar would show a stale count carried
over from the other tab. This is centralized in a `switchTab` helper used
by both the tabs' onValueChange and the "New session" snap-back (which
sets the tab outside Radix's onValueChange path), so no tab change can
skip the selection cleanup.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* Add product-analytics abstraction to web frontend
Introduce an opt-in, host-injected analytics seam so an embedding host can
collect user actions (clicks, field value-changes, page views) keyed by a
stable componentId. Fully inert standalone: when no host sink is configured
via OmnigentHostConfig.analytics, every emit is a no-op.
- lib/host.ts: OmnigentAnalyticsEvent type + analytics? sink + getter.
- lib/analytics.ts: emitOmnigentAnalytics, useOmnigentAnalytics
(trackClick/trackValueChange, values redacted by default for PII), and
useOmnigentPageView (re-fires on pathname change, like the unified router).
- Button/Input: optional componentId prop that reports clicks/value-changes.
- lib/routing.tsx: optional componentId on Link (OmnigentLinkProps) so a
link can opt into per-link analytics; standalone strips it.
- App.tsx: central <PageView id> wrapper declares each route's page-view id
next to the route table; SettingsPage keeps its own hook (param-derived
settings.<section> id) as the escape hatch.
- Example componentIds: chat composer send, tasks search, sidebar
conversation switcher, settings "Back to Omnigent" link.
Distinct from lib/telemetry.ts (low-level OTEL HTTP tracing); this is
application-level user-action analytics.
Co-authored-by: Isaac
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
* Ci
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
---------
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: Hubert Zub <hubert.zub@databricks.com>
* feat(sessions): auto-connect a wakeable runner on shell create
Creating a shell from the web UI on a session whose runner had gone to
sleep dead-ended on a 502 ("no runner available"), even though the host
was still up and the next chat message would have transparently woken it.
Add `ensure_runner_connected`, which runs the same runner-acquisition
ladder `post_event` uses (wake a stale resumable managed sandbox, launch
a runner on a live host, or relaunch a managed sandbox) without the
message-specific side effects, and call it from `create_session_terminal`
before proxying. Wakeable states reconnect and the shell opens; a
non-host-bound stranded session or an offline external host still 502s
(the CLI reconnect path owns those).
Surface connect state on the "+" → Shell menu item: it stays enabled and
shows "Reconnecting…" with a spinner while the server wakes the runner on
a wakeable session, and is disabled + labeled "Offline" for states the
browser can't reconnect. Widen the menu so the longer label isn't clipped.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(sessions): wait the connect grace before relaunching on shell create
Address PR review: ensure_runner_connected went straight to
_launch_runner_on_host whenever no runner client resolved, so opening a
shell against a session whose runner was merely booting (tunnel not yet
registered) would spawn a second runner and orphan the booting one —
diverging from post_event, which it claims to mirror.
When the session has a pinned runner_id and a live host, first wait
_HOST_BOUND_RUNNER_CONNECT_GRACE_S for it to connect (racing a
host.runner_status query that cuts the wait short if the host reports it
gone), and only relaunch if it's truly dead. Also drop the unused
tuple binding at the call site (the proxy re-resolves the client).
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(runner): allow managed mint in InitialAuthTokenFactory fallback
When a managed sandbox runner starts with a host-provided bearer
(_InitialAuthTokenFactory), and that bearer is rejected (401), the
fallback resolver was called with _allow_delegated_mint=False. This
blocked the managed-mint path entirely, leaving the runner with no
credential for its HTTP callbacks.
For managed runners (OMNIGENT_RUNNER_DELEGATED_AUTH=1 + binding token),
the fallback must be able to reach the managed-mint path after the
initial bearer expires — the same path used by runners that start
without a host bearer. Removing _allow_delegated_mint=False restores
this: SDK/OIDC auth still wins when present; managed mint is the
natural last resort for sandbox runners with no user credential.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): pass proxy bearer through managed mint so Apps proxy lets it through
The managed-mint endpoint (POST /v1/runners/{id}/token) is authenticated
by the runner's binding token, but on Databricks Apps deployments the
proxy layer sits in front and requires a valid Authorization header on
every request. With no bearer, the proxy returns 401 before the request
reaches Omnigent — the same symptom as the _allow_delegated_mint=False
regression, but a separate root cause.
Fix: thread an optional proxy_bearer through _make_managed_mint_factory,
_ManagedMintTokenFactory, and _mint_managed_owner_token, passed to
databricks_request_headers as the Authorization header. The initial
host bearer seeds it; after the first successful mint the minted JWT
replaces it as the proxy bearer for subsequent refreshes.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): reuse runner auth factory in codex discover-and-forward
_codex_discover_thread_and_forward was calling _make_auth_token_factory()
fresh, but RUNNER_INITIAL_AUTH_TOKEN is already popped from env by
runner startup — so the fresh call went straight to managed mint with no
proxy bearer, getting 401 from the Apps proxy before reaching Omnigent.
Fix: accept auth_token_factory at the call site, extracted from the
server_client's _RunnerDatabricksAuth (which already carries the correct
proxy bearer). supervise_forwarder also reuses it.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): store auth factory as singleton so all call sites share proxy bearer
Every _make_auth_token_factory() call after runner startup (harness setup,
terminal creation, forwarders) was building a fresh factory with no proxy
bearer, because RUNNER_INITIAL_AUTH_TOKEN had already been popped from env.
Each fresh factory hit the delegated-mint path, got 401 from the Apps proxy,
and left that call site with no credential.
Fix: store the factory built by serve_runner in a module-level singleton
(_runner_auth_factory). Subsequent _make_auth_token_factory() calls with
default args return it directly, so all call sites across orchestration.py
and app.py share the proxy bearer without any individual patching.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): fix __main__ vs omnigent.runner._entry module identity
When the runner runs as python -m omnigent runner, _entry.py executes
as __main__, creating a module object separate from omnigent.runner._entry.
Two bugs:
1. _runner_auth_factory was set on __main__ but read from
omnigent.runner._entry (always None). Fix: set it on the canonical
module via import omnigent.runner._entry as _self_module.
2. isinstance(server_client.auth, _RunnerDatabricksAuth) was False
because server_client.auth is __main__._RunnerDatabricksAuth while
the check used omnigent.runner._entry._RunnerDatabricksAuth. Fix:
use getattr(server_client.auth, _factory, None) instead.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor(runner): drop auth_token_factory param from codex discover-and-forward
Now that _make_auth_token_factory() returns the runner singleton (which
carries the proxy bearer), the explicit param and the server_client auth
introspection that fed it are no longer needed.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor(runner): introduce _set_runner_auth_factory to set singleton
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: use sys.modules to set singleton, restore docstring, remove dup comment
- Replace self-import with sys.modules lookup to avoid the module
importing itself (also sets on __main__ as a fallback).
- Move singleton early-return to after the docstring so __doc__ is
preserved on _make_auth_token_factory.
- Remove duplicated comment block in _codex_discover_thread_and_forward.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: import canonical module before setting singleton to ensure sys.modules registration
sys.modules.get() returns None when running as __main__ because the
canonical name isn't registered yet. Importing it first forces
registration, then both the canonical module and __main__ get the
singleton set.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: shorten overlong docstring in test
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: reuse singleton when server_url matches runner URL
Callers like native_policy_hook.py pass server_url explicitly but still
want the shared factory. The singleton guard now matches on both None
and the runner's own RUNNER_SERVER_URL.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
rendered.count("\n") undercounts when Rich wraps a long status label
(e.g. "✓ Isaac-Databricks-Ai-Gateway") across multiple terminal rows.
The cursor-up escape then doesn't move far enough, leaving stale menu
frames in the scrollback — which makes the "Configure harnesses" block
appear to stack on every loop iteration.
Replace the newline count with _count_terminal_lines(), which strips ANSI
escapes and uses ceiling division of each line's cell width by the terminal
width to count actual visual rows.
Tests cover no-wrap, wrapping, exactly-full-width, ANSI stripping, and the
empty-string edge case.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
The claude-native session's Working/idle badge is driven by diffing the
tmux pane (the PTY watcher in resource_registry). That heuristic can't
tell "blocked on a prompt" from "working", and only flips to idle after
~1s of pane quiescence rather than on the real turn edge.
Claude Code writes a per-process status file at
`<config_dir>/sessions/<pid>.json` (its internal "concurrentSessions"
registry, present since v2.1.139) whose `status` flips idle/busy/waiting
on the actual turn edges. Prefer that for the claude-native running/idle
status, falling back to the PTY watcher when the file is absent (old
Claude, missing config dir) or never resolves.
- New `omnigent/claude_native_status_file.py`: `resolve_status_file`
(pid-first via the tmux pane pid, which equals Claude's pid on this
launch path; sessionId cross-check + freshness-bounded scan fallback),
`read_session_status` (busy/waiting -> running, idle -> idle), and a
`SessionStatusPoller` that lazily resolves then mtime-polls the cached
path and emits deduped status edges, deactivating when the file
vanishes on clean exit.
- terminal.py: add `pane_pid_sync()` and an `on_tick` hook so the poller
runs on the existing watcher cadence — no second thread.
- resource_registry.py: for the claude-native role only, build the poller
and drive it via `on_tick`; while it is active the PTY on_activity/
on_idle edges defer status to the file. The PTY watcher keeps owning
the activity badge and exit detection, and reclaims status if the file
never resolves or disappears.
waiting maps to running for now (no new status vocabulary); surfacing a
distinct "needs input" state is a possible fast-follow.
Co-authored-by: Isaac
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
Update the Discord-watch roster (rotation_roster.json), leaving 9 people in the rotation. Prune elapsed dates from the schedule and extend the
horizon through 2026-10-30 so every upcoming weekday is assigned to a
current roster member.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): tune sidebar vertical spacing rhythm
Refine the sidebar's padding and gaps so the primary nav reads as a
proper section and the row lists sit on a consistent rhythm:
- Primary nav (New session / Automations / Inbox): 8px gap to the
Omnigent header (pt-2), no bottom padding of its own (pb-0); the 16px
gap below now comes from the scrolling list (pt-4), matching the
section-to-section gap-4 rhythm.
- Nav rows and session rows are 32px tall (h-8) with 4px vertical
padding (py-1).
- Section headers (Pinned / Projects / Sessions) get 8px bottom
padding (pb-2).
- Session rows and project folder rows stack flush (gap-0).
- Bulk-action bar uses uniform 6px padding (p-1.5).
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e-ui): regenerate visual baselines
* test(web): update sidebar spacing assertions to new rhythm
Bring the existing layout assertions in line with the tuned spacing:
primary nav pt-2/pb-0, nav + session rows h-8, iconless section header
pb-2.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e-ui): expect 32px session row height after spacing bump
Session rows moved from h-7 (28px) to h-8 (32px) in the sidebar
spacing tune; update the row-layout e2e assertion to match.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
When "Select sessions" is toggled on, the currently-viewed session's row
kept its active background even though it wasn't explicitly selected,
making the selection state ambiguous. Gate the active-route highlight on
`!selectionMode` so a row shows a background only when it's the active
session (normal mode) or explicitly checked (selection mode).
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): file new-in-project sessions under their project immediately
Stamp the omni_project label at session create so a session created from
the new-session composer is born filed under its project, instead of
flashing under the ungrouped "Sessions" section for a couple of seconds
until the follow-up project_id move catches up in the search-indexed
session list. The sidebar dual-reads project membership from the label
or the first-class project_id, so the row groups under its project from
its first appearance; the existing move then promotes it to project_id.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* test(e2e_ui): cover born-filed new-session-in-project flow
Add a Playwright e2e that lands on the /?project=<name> composer and asserts
the create POST /v1/sessions carries the omni_project label, so a session
created inside a project is filed under it immediately (satisfies the
E2E UI Required coverage gate for this web behavior change).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* docs(web): correct the born-filed move-failure catch comment
If the project_id move fails, the session stays filed via its create-time
omni_project label (not unfiled) — fix the stale catch comment.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* perf(web): make add-to-project instant — optimistic move + slim PATCH
Moving a session into a project waited on resolve→PATCH→refetch, with
the PATCH shipping a ~415KB items snapshot, so the row sat in its old
section for seconds. Overlay the membership optimistically from the
cached project id, render folder bodies as the union of their own pages
and the loaded window (so the row lands in-folder in one frame), and
return the PATCH snapshot without items (~1KB).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* chore(server): regenerate openapi.json for the PATCH sessions docstring
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): keep folder-only rows visible through an optimistic move
A row loaded only via an expanded folder's own pagination has no copy in
the flat window for the folder union to re-home, so dropping it from its
source folder blanked it from the sidebar until the refetches landed.
Insert such rows into the target folder's cached page and skip the
removal when nothing else can show them. Adds a browser e2e covering the
sidebar move flow end-to-end.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
Follow-up polish on the grouped/colored help (#3795). Focused on the
one-liner copy and a duplicate listing entry.
- Normalize harness short help to `Launch <Name> with Omnigent.` — was
an inconsistent mix of `Launch [the] <Name> [TUI] in an Omnigent
terminal`, and "in an Omnigent terminal" was noisy.
- Trim over-long / over-specific one-liners:
- `attach`: drop the "— never starts anything" clause (the body still
explains it's a pure client).
- `uninstall`: `Uninstall Omnigent from this machine.`
- `usage`: `Show your Omnigent usage and costs.` (was pinned to
today / 7 / 30 days).
- `upgrade`: `Upgrade Omnigent to the latest release.`
- `debug`: `Internal maintenance commands.`
- Hide the `update` alias (same Click object as `upgrade`) from the
listing via `_ALIAS_COMMANDS`, so it no longer shows as a duplicate
line; it stays registered and runnable.
- Update/extend tests for the new copy and the hidden `update` alias.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
Split the top-level `omnigent --help` command list into two sections —
`Harnesses` (agent/harness launch commands) and `Commands` (everything
else) — add brand-accent color, and hide harnesses whose optional extra
isn't installed (with a small notice pointing at `omnigent setup`).
- Add a `format_commands` override on `_OmnigentCLI` that partitions
visible subcommands using a `_HARNESS_COMMANDS` set, sharing one
aligned help column across both sections.
- Colorize headings (`Usage:`, `Options`, `Harnesses`, `Commands`) in
the brand accent, harness names in accent, other command names in
cyan, and option flags in green — via `format_usage`/`format_options`
overrides and a `_help_style` helper.
- Hide extras-gated harnesses (`cursor`, `antigravity`) from the listing
when their SDK isn't importable, via `_harness_extra_checks` (lazy
`find_spec` predicates). The commands stay runnable — running one
offers to install the extra. When any are hidden, show a dim notice
pointing at `omnigent setup` (which lists those harnesses and offers
the install), rather than enumerating extras that may change.
- Color is gated on `NO_COLOR` and Click strips ANSI on non-TTY sinks,
so piped/CI help stays plain. Alignment is ANSI-safe (Click's
`term_len` strips escapes before measuring columns).
- Add tests covering grouping, the extras-gated show/hide, and the notice.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
## Related issue
N/A
## Summary
- Added `--extra`, `--target-version`, and `--dry-run` flags to `omni upgrade`.
- Upgrade commands for `uv tool` and `pipx` now read the originally requested extras from the installer's receipt/metadata and preserve them.
- Explicitly refuses auto-upgrade for `pip` and `uv pip` because those installers do not record requested extras, making a safe automatic upgrade impossible.
- Fixed installer metadata detection in `uv tool` installs by avoiding `Path.resolve()` on the `bin/python` symlink, which previously pointed to the shared uv interpreter and missed `uv-receipt.toml`.
- Added/updated unit and CLI tests covering the new behavior.
## Test Plan
- `uv run pytest tests/cli/test_upgrade_command.py tests/cli/test_update_check.py tests/cli/test_cli.py -q --timeout=60` → **383 passed**.
- `uv run pytest tests/cli/test_update_check.py -q --timeout=60` → **112 passed**.
- Manually built a local wheel, installed it as a `uv tool`, and verified dry-run output.
- Verified `--extra` unions with detected extras.
- Verified `uv pip` install is refused with a manual-upgrade message.
## Demo
N/A
## Type of change
- [x] Feature
- [ ] Bug fix
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Manual verification was done by building local wheels and installing the current dev version as a `uv tool`:
```bash
# 1. Build wheels
rm -rf /tmp/omnibuild && mkdir -p /tmp/omnibuild
uv build --wheel -o /tmp/omnibuild .
uv build --wheel -o /tmp/omnibuild sdks/python-client
uv build --wheel -o /tmp/omnibuild sdks/ui
# 2. Install as uv tool with the "all" extra
rm -rf /tmp/omni-dev-test
UV_TOOL_DIR=/tmp/omni-dev-test uv tool install --find-links /tmp/omnibuild \
'/tmp/omnibuild/omnigent-0.8.0.dev0-py3-none-any.whl[all]' --force
# 3. Dry-run upgrade from outside the source repo
cd /tmp && /tmp/omni-dev-test/omnigent/bin/omni upgrade --dry-run --target-version 0.8.0
```
Output:
```text
Targeting v0.8.0.
Detected installer: uv
Detected extras: all
Would run: uv tool install --reinstall omnigent==0.8.0[all]
```
Adding `--extra server` unions with the detected extra:
```bash
cd /tmp && /tmp/omni-dev-test/omnigent/bin/omni upgrade --dry-run --target-version 0.8.0 --extra server
```
Output:
```text
Targeting v0.8.0.
Detected installer: uv
Detected extras: server
Would run: uv tool install --reinstall omnigent==0.8.0[all,server]
```
A `uv pip` install is correctly refused:
```text
omnigent was installed with `uv pip`, not `uv tool install`. `uv pip` does not record which extras were requested, so `omni upgrade` cannot preserve them safely. Upgrade manually:
uv pip install -U omnigent
# or, if you need extras:
uv pip install -U 'omnigent[your,extras,here]'
```
## Changelog
`omni upgrade` now preserves requested extras for `uv tool` and `pipx` installs, supports `--extra`, `--target-version`, and `--dry-run`, and refuses to auto-upgrade `pip` / `uv pip` installs because those installers don't record extras.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
## Related issue
N/A
## Summary
- `nodeLinker: hoisted` was a compatibility shim from the npm→pnpm migration
that forced an npm-style flat `node_modules`. Removing it returns pnpm to its
default isolated/symlinked layout (packages under `node_modules/.pnpm/…`),
restoring strict dependency isolation — dependencies must be declared, so
phantom/undeclared deps stop resolving by accident.
- Validated that the blockers the shim was assumed to guard against don't
actually block under the isolated layout (details in Test Plan). The Shiki
cyclic-import crash is handled by the existing `manualChunks` guard in
`web/vite.config.ts` (a chunking concern, independent of the node linker), and
electron-builder v26 collects the production dependency tree correctly through
pnpm's symlinks.
## Test Plan
Validated locally under the isolated layout:
- `pnpm install --frozen-lockfile` — clean and lockfile-consistent (the linker
setting is not part of the lockfile, so no lockfile churn).
- `pnpm --filter web run build` — succeeds; Shiki resolves to a single acyclic
chunk via the existing `manualChunks` guard.
- Electron packaging: `pnpm --filter web run build:overlay` then
`electron-builder --dir` builds and signs the app; inspected the resulting
`app.asar` — it bundles exactly the production dep tree (`electron-updater`,
`js-yaml` + their 14 transitive deps) with zero dev-dependency bloat.
- Tailwind v4 `@source` scan follows the symlink: the emitted CSS is
byte-identical between the hoisted and isolated builds.
- oxlint (schema) and prettier run; `node web/node_modules/vite/bin/vite.js
--version` (Android Gradle entry) and `web/node_modules/.bin/tsc --version`
(iOS Fastlane probe) resolve via pnpm's direct-dependency symlinks.
Not runnable locally — relying on CI to confirm: Docker image build,
`electron-build` full installers, and `android-bundle` / iOS app builds.
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [x] Not applicable
## Coverage notes
Node-linker layout has no unit-test surface; verified manually via a full
install + web build + electron `--dir` packaging (inspecting the packaged
`app.asar` dependency tree) + a Tailwind CSS byte-diff, and confirmed the
hardcoded node_modules paths (vite entry, tsc/prettier/oxlint) resolve through
pnpm's direct-dependency symlinks. Remaining platform builds are covered by CI.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
N/A
- Remove the legacy `_LAUNCHERS` fallback from `__init__.py` — all providers
are now resolved exclusively through the `SandboxProviderRegistry`
contribution-based registry.
- Simplify `get_launcher()` to a single code path (no more
`DeprecationWarning` / legacy import fallback).
- Remove unused `warnings` / `importlib` / `importlib.util` imports from
`__init__.py`.
- Add `docs/extending/sandbox_providers.md` documenting how to implement and
register a third-party sandbox provider, including a minimal example
package with `pyproject.toml` entrypoint, the namespace requirement, and
the capability reference table.
```bash
uv run pytest tests/onboarding/sandboxes tests/sandbox tests/cli/test_cli.py tests/server -k managed -q
pre-commit run --files <changed files>
```
All 782 selected tests pass and pre-commit is clean.
N/A
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [ ] Manual verification completed
- [x] Existing tests cover this change
- [ ] Not applicable
Existing tests pass unchanged. The test that expected a `DeprecationWarning`
from the legacy path was updated to no longer suppress it. New docs are
prose-only and need no test coverage.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
## Related issue
N/A
## Summary
- The `manualChunks` guard that keeps Shiki in one chunk (to avoid a cyclic
import crash — the language-index ↔ alias-map split that throws "Cannot read
properties of undefined (reading 'flatMap')" and blanks the Monaco/file
viewer) matched both `/shiki` and `/@shikijs/`. That also swept every
`@shikijs/langs/<lang>` grammar — which Shiki loads via dynamic import as
per-language chunks — into the single, eagerly `modulepreload`ed core chunk.
So ~200 language grammars (~1.68 MB gzip) were downloaded on every initial
page load, even though a session uses only a few languages.
- Exclude `@shikijs/langs/<lang>` from the `shiki` chunk so grammars stay lazy
per-language chunks. Keep Shiki's core, engines, and bundle glue together so
the cyclic core stays intra-chunk — the engines must stay too: excluding them
re-splits the cycle across chunks and reintroduces the `flatMap` crash.
- Initial-load eager JS drops from ~11.8 MB to ~4.37 MB (Shiki 1.68 MB → 466 KB
gzip); grammars become 427 on-demand chunks. Layout-independent (same result
under pnpm hoisted and isolated).
## Test Plan
- `pnpm --filter web run build` succeeds.
- Verified the emitted `shiki` chunk statically imports only the rolldown
runtime (no cross-chunk cycle) and contains `bundledLanguagesAlias`
co-located with its reader — under both hoisted and isolated node_modules.
- Verified per-language grammar chunks (python, rust, typescript, …) are
emitted separately and are NOT `modulepreload`ed by `index.html`.
- Recommended pre-merge smoke test: open the file viewer / Monaco editor and a
markdown code block and confirm syntax highlighting renders.
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [x] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Verified via build output analysis: the `shiki` core chunk is acyclic with the
alias map co-located (cycle fix preserved), and language grammars are emitted as
separate, non-preloaded chunks. Existing Shiki/code-block tests exercise the
runtime highlighting path; this change only affects chunk grouping, not module
behavior.
## Changelog
Faster initial load — syntax-highlighting language grammars now load on demand instead of all upfront.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
Remove the release-specific Gemini fallback from the Antigravity SDK executor. Preserve explicit per-turn and HARNESS_ANTIGRAVITY_MODEL precedence, but omit LocalAgentConfig.model when neither is set so every supported google-antigravity 0.1.x release owns its current default for both API-key and Vertex sessions.
Expand the no-hardcoded-model scanner to recognize dotted, canonical, and normalized Gemini release ids. Add coverage that distinguishes an omitted SDK model from an explicit override, and remove the stale release example from runtime error text.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
The embed build set sourcemap: false, so downstream bundlers that embed this
output (e.g. the Databricks monolith's rspack/webpack) had no input map to
chain through — host-side error stack frames bottomed out at
omnigent-embed.js:<line> instead of the original src/**.
Emit maps so the embedding host can compose them to source. dist-embed is a
build artifact (gitignored), so this ships nothing new; it only enriches the
maps hosts consume via source-map-loader.
Signed-off-by: Hubert Zub <hubert.zub@databricks.com>
Co-authored-by: Hubert Zub <hubert.zub@databricks.com>
* feat(sandbox): scan write_paths for dotfiles too
The dotfile / escaping-symlink masker walked cwd and every read_paths
root but skipped write_paths, so a writable directory granted outside
cwd could still leak — and let the helper overwrite — top-level secrets
like .env / .aws / .ssh.
Fold read_paths and write_paths into one deduplicated, ancestor-first
set via a new merge_scan_roots helper so every granted root is masked,
and a path granted by more than one lever (or nested under another
grant) is walked once instead of once per lever. The dedup resolves
each root a single time and skips nested roots with a lexicographic
cover scan, so the big-grant profile-size guard stays fast.
Signed-off-by: Thomas Garnier <6202935+mxatone@users.noreply.github.com>
* fix(sandbox): only drop nested grants when scanning recursively
Review caught that merge_scan_roots dropped a granted root whenever
another grant was its ancestor — but that subsumption only holds when
the walk is recursive. cwd_hidden_scan_recursive defaults to False,
where each walk masks only a root's immediate children, so dropping a
nested grant (e.g. write_paths: [/a/deep/nested] under read_paths:
[/a]) left its top-level dotfiles visible and writable — reintroducing
the exact leak this branch closes, and regressing the prior
per-read-root behavior.
Thread the recursive flag into merge_scan_roots: keep the cwd drop
(unchanged, pre-existing), but only collapse a grant into a kept
ancestor when recursive=True; in top-level-only mode keep every
distinct grant and drop only exact duplicates. Walk the full ancestor
chain (not just the last kept root) so an interleaving sibling name
cannot hide a real ancestor and leave a redundant walk.
Adds regression tests in both backends for the non-recursive nested
grant, plus merge_scan_roots unit coverage for both modes.
Signed-off-by: Thomas Garnier <6202935+mxatone@users.noreply.github.com>
* fix(sandbox): exclude framework scratch roots from the write-root dotfile scan
Extending the dotfile mask scan to write_paths also swept in the
framework-added scratch tmpdir (folded into write_roots via
with_additional_write_roots). That dir holds the sandbox's own egress
relay socket `.egress.sock` — a dotfile — so the scan masked it with
`--bind-try /dev/null` (bwrap) / a deny rule (seatbelt), cutting the
relay endpoint and resetting every egress connection. This is what broke
the inner-rest `test_egress_e2e[linux_bwrap]` cases.
Track framework write roots on the policy as `mask_scan_skip_roots` and
drop them (and anything nested under them) from `merge_scan_roots`. These
dirs are created fresh by the framework and never hold pre-existing user
secrets, so scanning them is both pointless and harmful. Genuine
user-declared read/write grants are still scanned.
Signed-off-by: Thomas Garnier <6202935+mxatone@users.noreply.github.com>
---------
Signed-off-by: Thomas Garnier <6202935+mxatone@users.noreply.github.com>
* lint(models): remove the hardcode baseline
Delete the empty path/count allowlist and its parser, stale-count logic, tests, and special pre-commit trigger. The scanner now rejects every non-owned production model literal while retaining only the AST-verified StaticModelFallback boundary.
Update the migration plan to describe the final configuration/catalog/fallback state. The fully merged issue 3426 audit passes 136 focused tests, mypy, the hardcode scan, and full pre-commit.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* lint(models): scan every production model literal
Remove the model-context name heuristic so positional arguments, bare collections, and config values under arbitrary keys cannot bypass the hardcoded-model check. Preserve docstrings and structurally owned fallback records as explicit non-runtime exceptions, and distinguish complete model ids from stable family-prefix compatibility checks.
Curate the newly exposed production literals by resolving Claude's direct-login custom model through the central owned fallback and replacing release-specific CLI, Bedrock, and loader examples with provider-neutral guidance.
Validated with 172 lint/Claude tests, 63 loader tests, focused mypy, the baseline-free repository scan, and pre-commit run --all-files.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* lint(models): cover the full production tree
Run the hardcoded-model scanner across every tracked Python and supported config/shell file rather than a curated directory list. Exclude tests and generated OpenAPI explicitly, and keep the pre-commit trigger exactly aligned with the scanner surface.
Remove the unused root server config that pinned a stale Databricks model and profile. A repository-wide dry run found no other non-generated production literals outside the existing scan surface.
Validated with the focused lint suite, the baseline-free full repository scan, focused mypy, and pre-commit run --all-files.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): keep Claude custom fallback routable
Resolve the private Sonnet custom-picker id through the exact owned fallback when available, then through the first routable Sonnet-family entry if release naming drifts. Fail clearly when the owned subscription catalog contains no Sonnet entry instead of forwarding an invalid picker id.\n\nRemove the vestigial full-tree scan-root constant, keep the pre-commit parity probe direct, make the Gemini docstring fixture exercise a recognized id shape, and update the migration guide to describe the actual full-tree literal scan and runtime-prose expectations.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* ci(models): configure automation model roles
Replace provider release ids in credentialed workflows with six repository-variable roles covering Anthropic, fast Anthropic, OpenAI, E2E judge, E2E model pool, and image generation workloads.
Make the shared Omnigent agent action require an explicit model input, validate required configuration before writing provider files, and keep fail-open reviewer/image helpers on their existing degradation paths.
Use the protocol-level mock-model fixture for mock-only integration matrices and remove their unused production model-spread configuration.
Validation: parsed all action/workflow YAML; generated integration and backcompat matrices; hardcode lint and staged pre-commit passed.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* ci: fail fast without E2E judge model
Require the repository-level E2E judge model variable before running the required-check script. This turns an absent CI configuration into an immediate, actionable failure instead of allowing a later command to fail ambiguously.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* ci: clarify missing optional model variables
Keep the advisory reviewer ranker, VS Code changelog drafter, and feature-blog image generator fail-open when their repository model variables are empty.\n\nEmit actionable variable names before skipping or falling through to the existing warning path, avoiding malformed gateway requests while preserving the best-effort behavior of all three jobs.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Add a lint step that fails when a `pnpm-workspace.yaml` `overrides:` pin
doesn't satisfy the range the same package declares in a workspace
`package.json`. Overrides outrank package.json, so such a mismatch
silently ignores the declared version — the trap that let a postcss
security bump (`^8.5.18`) land while the override still pinned the
vulnerable `8.5.15`, invisible to both `--frozen-lockfile` and the
lockfile-regen gate (the lock was internally consistent for the pin).
Runs in the lint job beside the existing "Check pnpm-lock.yaml is up to
date" step. The checker uses a small npm-flavored semver comparison
(`^`, `~`, exact, comparators) over the operators this repo uses;
unrecognized ranges are reported rather than passed silently.
Co-authored-by: Isaac
* ci(release): add source-PR demo-video table to release-post PRs
The publish-changelog workflow reformats a published release into a site post
that leaves a `TODO` demo placeholder under each feature, with no pointer to
the source PRs that may already ship a recording. Parse the feature PR refs
from the curated release body (Major new features / Breaking changes sections;
bug fixes are dropped from the post, so from the table too), detect whether
each PR already has a demo video attached (same detection as feature-blog.yml
— uploaded asset links, bare .mp4/.mov/.webm/.m4v URLs, <video> tags; images
not counted), and inject a per-section PR | Title | Demo video? table into the
release-post PR body (and the dry-run preview) so reviewers can drop an
existing clip into a placeholder instead of re-recording.
Runs independent of the LLM reflow so it also helps the raw-body fallback;
best-effort (continue-on-error), leaving the table empty on failure.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* ci(release): group demo-video table by the post's curated features
The demo-video table grouped PRs by the raw release-body sections, so it listed
every feature PR (e.g. all 20 under "Major new features") even though the
published post is curated down to a handful of headline features, each with one
demo placeholder. Reviewers saw far more PRs than the post has slots for.
Have the release-post-formatter emit a RELEASE_POST_PRS map (feature title ->
contributing PR refs) after the post, and build the table from that so its
groups match the post's numbered features and only list the PRs behind them.
Validate the map against the harvested PR set. When no map is present (raw-body
fallback, where the post keeps every feature), fall back to grouping by the raw
release sections as before.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* ci(release): match feature-section headings at any level
The demo-table section parser matched only `## ` headings, but the release body
uses `### ` (h3) section headings, so it found zero feature sections and built
an empty table. Match `#{2,}` and test the heading TEXT with startswith, so
"Major new features" / "Breaking changes" match at any level while "Bug fixes
& hardening" and "Thanks to our community" are still excluded.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
The pnpm-workspace.yaml `overrides:` block force-pinned postcss to
8.5.15 and linkify-it to 5.0.1 — both flagged by open high-severity
advisories (postcss GHSA-r28c-9q8g-f849, linkify-it
GHSA-v245-v573-v5vm). Because the override sits above package.json, the
earlier dependabot bump of postcss to ^8.5.18 (#3385) was inert: the
lock kept resolving 8.5.15, so the CVE was never actually fixed, and the
frozen-lockfile gate saw no drift.
Bump the two override pins to the patched releases and regenerate the
lock (postcss 8.5.18, linkify-it 5.0.2). Both are same-minor patch
bumps confined to security/bug fixes — unlike the vite/tailwind/
lightningcss pins in the same block, they aren't the bundler, so they
don't affect chunk splitting or the Shiki/PDF-worker asset emission the
override comment warns about. CI's Docker build + web test validate the
bundle.
Co-authored-by: Isaac
* feat(models): persist last-known-good catalogs
Persist validated MLflow provider catalogs under the platform user-cache directory so catalog-backed defaults survive transient GitHub and release-CDN outages after one successful fetch.
Keep the existing one-hour freshness window, fall back to stale validated data for at most seven days after a live failure, and record cache schema, upstream schema, source URL, and fetch time. Atomic replacement keeps concurrent writers from exposing partial JSON, while corrupt, incompatible, wrong-source, and over-age entries fail closed.
Make OMNIGENT_DISABLE_CATALOG_LOOKUP bypass memory, disk, and network state for hermetic tests. Cover persistence, fresh reuse, stale provenance logging, corruption repair, schema/source rejection, over-age behavior, concurrent writes, and first-run failure.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): accept compatible catalog schemas
Validate the current MLflow catalog shape without coupling live discovery or persistent cache reuse to one exact minor schema string. Accept major-version-compatible string and integer forms, continue rejecting unsupported majors and malformed values, and document when stale in-memory fallbacks retry discovery.
Production release assets for Anthropic, OpenAI, Gemini, and OpenRouter were verified against the validator; focused catalog tests and full pre-commit pass.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): preserve cache across empty catalogs
Reject empty live catalog model maps so transient or truncated upstream payloads cannot overwrite useful last-known-good data. Tighten compatible schema parsing to ASCII digits so corrupt cache metadata is ignored instead of raising.
Percent-encode provider names in release asset URLs to keep path and query delimiters inert. Add regression coverage for empty-result fallback preservation, non-ASCII schema corruption, and URL construction.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* ci(oss): gate lockfile regen on a consistency check
The OSS lockfile-regen job deleted pnpm-lock.yaml and re-resolved from
scratch every 12h, so any in-range transitive drift on public npm
produced a ~1500-line churn PR that advanced ~20 unreviewed deps for no
functional reason (e.g. #3631). The job exists to keep the tree
Docker-buildable when a manifest change desyncs the lock — not to chase
newer upstream versions.
Add a check-first gate: `uv lock --check` and pnpm
`--frozen-lockfile --lockfile-only` verify each lock still satisfies its
manifests. These pass on a consistent-but-not-latest lock, so routine
drift no longer triggers a regen; only a real manifest/lock desync flips
`drifted=true` and runs the regenerate → Docker smoke → PR steps.
Also correct the PR-body text, which claimed it regenerated "uv.lock +
web/package-lock.json" (the repo locks pnpm-lock.yaml, not
package-lock.json).
Co-authored-by: Isaac
* ci(oss): keep the Docker smoke on the no-drift path
Per PR review: ungate the Docker build + CLI smoke so they run every 12h
regardless of drift. On the drifted path they still validate the freshly
regenerated locks before commit; on the clean path they remain the
ongoing proof that the committed locks + public registries build a
working image — catching buildability regressions independent of
manifest state (a yanked-but-in-range package, a Dockerfile break) that
the check-only gate would otherwise miss.
Co-authored-by: Isaac
* ci(oss): gate each ecosystem's regen on its own drift flag
Per PR review: a single shared `drifted` flag meant a desync in one
ecosystem (say uv.lock) still ran the `rm -f pnpm-lock.yaml &&
pnpm install` from-scratch regen of the other, re-resolving it against
public npm and reintroducing exactly the in-range transitive churn this
job is meant to avoid.
Split into `drifted_uv` / `drifted_pnpm` and gate each Regenerate step
on its own flag. A combined `drifted` (either) still drives the shared
token-mint and open-PR steps; the commit stages only whichever lockfile
actually changed.
Co-authored-by: Isaac
* feat(web): redesign sidebar bulk-selection bar and scope it per section
Rework the sidebar's bulk-selection UI into a single bordered "pill" bar
rendered directly under the header of the section it targets, and give
selection an explicit scope so it acts on the right rows.
- Bar redesign: one pill row with an Exit (X) button, an "N selected"
count at the session-title font size, and icon-only Archive + Delete
actions. Archive shows by default and is disabled until an archivable
session is selected (Delete likewise). Unarchive replaces Archive only
when the selection is entirely archived.
- Row checkbox moved to the left of the session title.
- Selection scope: the Sessions-header trigger selects the flat session
list; the Projects-header kebab's "Select sessions" selects the
sessions nested inside project folders (bar renders under the Projects
header). Entering a scope preserves current folder expansion. The
shift-select range and a stranding guard follow the active scope.
- Fold the Projects expand-all/collapse controls plus "Select sessions"
into a kebab to the right of the New-project (+) button.
Test-only: update unit tests for the new layout/scoping and rewrite the
e2e-ui bulk-actions suite (5 passing) to match the redesign, including a
projects-scope round-trip.
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): resolve projects-scope selection against folders' own rows
Projects-scope bulk selection sourced its action set, shift-select range,
and stranding guard from the global paginated window
(sections.projectGroups), but each ProjectFolder renders from its own
independent useProjectSessions query. A folder member outside the global
window would toggle the count yet silently drop from bulk archive/delete,
break shift-select, or trip the stranding guard.
Each ProjectFolder now reports its rendered rows up via
onConversationsLoaded; the parent unions them (deduped) into a
projectSessionPool that backs the bulk-action bar, the shift-select range,
and the guard — so all three agree on what's selectable regardless of the
global pagination window.
Adds a regression test: with the folder query returning p1,p2,p3 while the
global window holds only p1,p2, shift-select p1->p3 spans all three and
bulk-archive fires with p3 included.
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): surface owned Delete count and guard selection-mode against transient empties
Addresses two non-blocking review notes on the bulk-selection bar:
- Delete acts only on owned rows, so a mixed-ownership selection (reachable
in projects scope, where a folder can hold others' sessions) read
"N selected" while Delete hit fewer. The Delete control's label/tooltip
now shows the owned count ("Delete 2") when it differs from the selection
size. Archive needs no such hint (its enable-gate already forces a
uniform archive group, and archived rows never appear in a selectable
section).
- The stranding guard that exits selection mode when the pool empties now
skips while the sessions query is refetching, so a background refetch
that briefly yields an empty page can't kick the user out mid-task.
Adds a mixed-ownership Delete-label test and updates the layout spec's
label assertion.
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* docs(web): clarify projects-scope stranding guard can't exit on a folder refetch
The exit-on-empty guard suppresses the global query's refetch via
conversationsQuery.isFetching, but the projects pool is fed by per-folder
queries too. Note that the pool unions global-derived membership, so a
single folder's transient-empty refetch can't zero it while any member is
in the global window — only a genuinely empty pool exits. Comment-only.
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
The feature-blog workflow leaves a `DEMO REQUIRED` marker in each drafted
post and tells the reviewer to record a demo, with no hint that the source
PRs may already ship one. Collect the contributing PRs per feature and detect
whether each already has a demo video attached (uploaded asset links, bare
.mp4/.mov/.webm/.m4v URLs, or <video> tags — images are not counted), then
inject a PR | Title | Demo video? table into the draft PR body so reviewers
can pull an existing recording into the marker instead of re-recording.
Reuses the gh pr view call already made to pick the reviewer (extended with
title/body/url). The table is written per feature even when no PR has a video.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* feat(web): rework workspace rail — tabs, shell-as-tab, maximize, + menu
Reworks the desktop right "Workspace" rail's tab strip so open items and
navigation read as one editor-style set, and gives shells a home inside the
rail instead of taking over the chat column.
- Reorder the strip: open file/shell tabs own the flexible left region; the
static nav tabs (Files/Agents/Shells/Tasks/Browser) sit right when tabs are
open, else stay anchored left.
- Shells open as top-strip tabs (desktop): clicking a shell row opens it as a
closable rail tab whose xterm renders in the rail's content slot — the chat
page is undisturbed. Mobile keeps the full-screen drawer.
- Add a full-screen (maximize) toggle pinned to the rightmost edge; maximized
keeps the docked card styling (same inset/height), only the width changes.
- Add a "+" menu ("Open new" → Shell) that trails the last tab when tabs are
open, else sits by the nav tabs. Browser stays a pinned tab (one embedded
WebContentsView per conversation).
- Tighten strip spacing and give the nav icons a consistent hover background;
smaller shell-tab label text.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): keep one ml-auto in the rail strip; drop phantom gap & maximize padding
Follow-up layout fixes to the workspace rail tab strip:
- Only ever one ml-auto in the strip row — two siblings both claiming it split
the free space and stranded the nav group mid-strip. With open tabs the
divider owns ml-auto (dragging nav + maximize right together); with no tabs
the maximize button owns it (nav group stays left).
- The divider dropped its ≥500px container-query gate so it shows at any rail
width instead of vanishing on a narrow rail.
- FileTabsStrip / TerminalTabsStrip return null when empty — an empty wrapper
still consumed a slot in the region's gap and left a phantom gap before the
trailing "+".
- Removed the maximize button's pl-0.5 so it sits flush like the other icons.
Adds regression tests asserting exactly one ml-auto per strip state, the
divider's presence/placement, the no-phantom-gap child count, and no maximize
padding.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): flat tab hover background — opaque fill, no gradient patch
The tab hover used bg-muted, but --muted is a translucent token (6% black).
The close-button overlay then faded in a second translucent gradient on top,
stacking alpha on the right edge into a visible darker patch. Use the same
opaque color-mix selection surface the active tab uses for both the hover
background and the overlay gradient, so hover is a flat even fill.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e): update shell-open tests for rail-tab behavior
Shells now open as tabs in the workspace rail (xterm in the rail content
slot) instead of taking over the main column via MainTerminalView. Update
the three e2e tests that asserted the old main-column flow:
- shells/test_new_shell: assert the shell opens as a rail tab (Close
"zsh · u-…" x + rail-scoped xterm) with the chat surface undisturbed.
- files/test_right_panel: clicking a shell row opens a "zsh · main" rail
tab; xterm connects in the rail, chat not replaced.
- sessions/test_terminal_theme: resolve the connected xterm inside the
Workspace rail rather than main-terminal-view.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e-ui): regenerate visual baselines
* feat(web): shell-type picker, pinned rail tab strip, sidebar restore
Follow-ups to the workspace-rail rework:
- "+" menu Shell entry: clicking Shell launches the remembered default type
immediately (selection optional); the submenu check-marks and remembers the
last-picked type (persisted to localStorage), used as the next default.
- Removed the Shells tab's "+ New shell" row — shell creation now lives solely
in the "+" menu. The Shells tab is a pure list.
- Hide the Shells tab (and mobile entry) unless a shell actually exists; merely
declaring shell access no longer surfaces an empty tab.
- Tab strip: nav icons + divider stay pinned left and the "+" stays pinned right
at every rail width — the tabs region is the sole horizontal scroller, and the
"+" sits outside it (no scroll/overlap). Divider shows at all widths again.
- Full screen: collapse the left sidebar on enter and restore its prior state on
exit (collapsed stays collapsed, open reopens).
Updated unit + e2e tests to match (shell-open via the "+" menu; Shells-tab gate).
Co-authored-by: Isaac
* fix(web): keep "+ New shell" in the mobile Shells drawer
Removing the "+ New shell" row broke first-shell creation on mobile, which has
no tab-strip "+" menu. Restore it there only:
- InlineTerminalsSection gains an opt-in ``showNewShell`` prop (default off);
the desktop rail stays list-only, the mobile drawer passes it to surface the
create row.
- The mobile Shells menu entry gates on existing-shell OR declared shell access
(so the drawer is reachable at zero shells), while the desktop rail tab stays
gated on an existing shell.
- Update the two e2e tests that opened a shell via the removed row to use the
"+" menu; the mobile drawer test's docstring clarifies the mobile-only create
path.
Co-authored-by: Isaac
* test(e2e-ui): regenerate visual baselines
* fix(web): restore sidebar on session-switch un-maximize; detangle toggle
Addresses Polly review notes on the full-screen sidebar handling:
- The session-switch reset un-maximizes the rail directly, but didn't restore
the sidebar it collapsed on entry — so maximize → switch conversation left the
sidebar silently collapsed. Extract restoreSidebarAfterMaximize() and call it
from the reset (only when we were maximized).
- Move the sidebar side effect out of the setRightPanelMaximized updater into a
plain toggleRightPanelMaximized handler, so the state setter stays a pure
prev→next flip instead of nesting other setters.
Co-authored-by: Isaac
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
* refactor(pi): discover inner gateway models live
Replace the seven-model Databricks registry embedded in the inner Pi executor with the workspace's Unity Catalog model-service listing.
Enrich live entries with MLflow context and output limits when available, while retaining the selected-model registration path so catalog outages do not prevent a configured session from launching.
Expose normalized max-output metadata, cover live routing and offline behavior, and ratchet all seven Pi entries out of the hardcode baseline.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(pi): normalize selected catalog aliases
Rewrite a live Unity Catalog alias to the exact configured Pi launch selector before rendering models.json. This keeps the menu deduplicated without dropping the concrete id Pi must resolve at startup.
Also document why explicit selections bypass picker compatibility filtering, remove a stale static-list reference, and make scalar metadata precedence explicit. Preserve live token metadata in the alias regression test.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
The first message is silently ignored (sandbox/lakebox wake) or
double-processed (managed relaunch) because of a race between the
server's persist-before-forward invariant and the runner's
crash-recovery turn detection.
When the server calls session-init (POST /runner/v1/sessions) immediately
before forwarding a message — managed sandbox wakes, sub-agent binding
repairs, host relaunches — the runner loads history during create_session.
Since the server already persisted the message to DB (invariant I1), the
runner sees it as a pending user message and starts a crash-recovery turn.
The subsequent message forward then arrives to an occupied _active_turns,
gets buffered, and is processed a second time once the recovery turn
finishes.
Add suppress_recovery_turn to the session-init envelope. The server sets
it True whenever it calls session-init as part of the message-forward
flow, so the runner skips recovery-turn detection and the forward is the
sole trigger for the turn.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* docs(models): remove stale runtime model examples
Describe Bedrock inference profiles, routing policy inputs, and child-session overrides in provider-neutral terms instead of recommending release-specific model ids in runtime help.
Ratchet the five corresponding hardcode-baseline entries and document that concrete examples belong in tests or provider-owned documentation, where they cannot become stale runtime guidance.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* docs(models): retain Bedrock id shape guidance
Keep the setup prompt provider-neutral while showing the non-obvious inference-profile identifier shape. The hint uses placeholders instead of a release-specific model id, so it remains useful without becoming stale or expanding the hardcode baseline.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* Document provider-neutral model id shapes
Restore useful model-format guidance with synthetic, non-release examples in Bedrock setup, routing policy, and child-session help. Keep concrete release ids out of runtime text so examples teach syntax without becoming stale recommendations.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Add an `allow_destructive` parameter (default `False`) to the GitHub
policy that separately gates irreversible destructive operations
(deletes). Normal writes (create, update, push) are still governed
by `write_repos` / `write_branches`; destructive operations require
BOTH being in `write_repos` AND `allow_destructive=True`.
Destructive operations gated:
- MCP: delete_file, delete_branch, delete_release
- Shell git: git push --delete, git push origin :branch
- Shell gh: delete actions across 13 groups (repo, release, issue,
gist, cache, codespace, project, variable, ssh-key, gpg-key,
secret, label, run)
For MCP, the destructive check fires after the repo allowlist so a
destructive op on a non-allowed repo still gets the repo DENY. For
shell ops, the destructive DENY fires early since even an
undeterminable-repo destructive op should be DENY not ASK.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(sessions): reject undeclared sub_agent_name at create (#3526)
POST /v1/sessions persisted an arbitrary `sub_agent_name` with no check
that the parent's spec declares it. Every downstream site that swaps in
the resolved child spec is guarded by `if ... is not None` with no
`else`, so a name that resolves to nothing left the parent spec, workdir,
harness and instructions in place — silently booting the child as a full
clone of the parent (runaway recursion for an orchestrator), with nothing
logged and nothing failing.
Fail loud at the create route: `_require_declared_subagent` loads the
trusted parent bundle and rejects a name the spec does not declare with
404, before any row is persisted. This mirrors normal `sys_session_send`
dispatch and the AGENTSPEC.md contract that unlisted names are rejected.
The check only fires when the bundle loads and the name is positively
absent; a load failure or absent cache cannot prove the negative and is
left to fail-loud downstream.
Defense-in-depth: the four runner spec-swap sites now log a warning on a
resolve-miss (`_warn_unresolved_sub_agent`) so stale rows or post-create
bundle edits that still reach the fallback are diagnosable instead of
invisible.
Test: test_subagent_create_rejects_undeclared_name asserts the create
route 404s on an undeclared name.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* test: declare sub-agents in tests that create children (#3526)
The new create-time gate rejects a `sub_agent_name` the parent spec
doesn't declare, which broke existing tests that spawned children of a
sub-agent-less parent:
- test_sessions_endpoints.py: two external-status tests created a
`worker` child of the default (no-sub-agent) agent. `create_test_agent`
now takes `sub_agents`; both declare `worker`. `build_agent_bundle`
gives each bundled sub-agent a default `claude-sdk` harness (the strict
spec_version:1 parser requires one for an omnigent executor).
- e2e_ui/conftest.py: the `hello_world` fixture now declares a
`researcher` sub-agent inline, so the mobile-workflow and
subagent-tab-title fixtures can spawn a `researcher` child.
Full tests/server/integration/ suite passes (995 passed, 3 xfailed).
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
The changelog on main skipped from v0.5.0 to v0.7.0, missing both
released tags. Backfill v0.6.0 and v0.5.1 in version order, and remove
the orphaned [Unreleased] block (its two entries — the Nord theme #2561
and per-harness command overrides #2933 — are already covered by the
v0.6.0 section).
v0.6.0 entries are cleaned from the auto-drafted PR #2960: dropped
non-entries (placeholder "written by Isaac" lines, "DELETE THIS SECTION"
markers, N/A refactor/cleanup notes), de-duplicated entries already
recorded under v0.5.0 (#1835, #2371), and normalized doubled tag
prefixes. v0.5.1 is from PR #2395.
Supersedes and closes#1843, #1897, #2395, #2960.
Co-authored-by: Isaac
* fix(runner): evaluate PHASE_TOOL_CALL policy for sys_call_async background tasks
Out-of-turn sys_call_async dispatches run in a detached asyncio task after
the originating turn ends. The executor adapter's _stable_policy_evaluator
reads _current_ctx which is cleared to None by run_turn's finally block, so
PHASE_TOOL_CALL evaluations always fail closed to DENY regardless of the
configured policy.
Fix by evaluating PHASE_TOOL_CALL directly via the AP server's REST endpoint
before executing the background tool. This bypasses the SSE round-trip (which
requires a live turn stream) and instead calls POST /sessions/{id}/policies/evaluate
inline from _bg(). ASK is treated as DENY since there is no active turn to
surface an approval prompt.
Sessions without a server_client or conversation_id (e.g. tests) skip
evaluation, preserving existing behavior.
Fixes#3233.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(runner): pass arguments as dict in async PHASE_TOOL_CALL evaluation
The initial commit sent target_args (a JSON-encoded string) as the
arguments field. Every other PHASE_TOOL_CALL evaluation path sends a
dict, and the server's policy context builder + built-in safety policies
(e.g. argument-aware rules that inspect arguments.command) expect a dict.
Sending a string caused isinstance(args, dict) checks to fail silently,
so argument-scoped DENY/ASK policies couldn't inspect the async tool's
arguments.
Parse target_args into a dict before building the evaluation body, with
a fallback to {} for malformed input. Add a test assertion that verifies
the forwarded arguments are a dict with the correct contents.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* docs(runner): clarify ASK parking behavior in async policy evaluator docstring
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(claude-sdk): stop claiming live message queue support
ClaudeSDKExecutor.enqueue_session_message() called query() which queues
a new turn on the SDK's stdin rather than injecting into the active turn.
Returning True from this method caused the adapter to emit
injection.consumed, dropping the runner's buffered copy. The next user
message would then trigger a turn with an empty buffer, answering the
previous message — producing a permanent one-turn-behind desync.
Fix: return False from both enqueue_session_message and
supports_live_message_queue. The adapter's existing if-not-accepted
branch retains the message and delivers it as a normal continuation turn
once the active turn ends, preserving in-order delivery.
Closes#3472.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(lint): suppress ARG002 for unused-but-required override params
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(claude-sdk): send all batched steered messages, not just the last
When a user steers multiple messages during a running SDK turn, each is
buffered and the runner collapses them into one continuation turn whose
history ends in several consecutive user messages. On a resumed SDK
session _build_prompt called _extract_latest_user_content, which walks
history in reverse and returns only the FIRST user message it finds — so
the SDK saw just the last steered message and the earlier ones were
silently dropped (they remained in the transcript, making it look like
the second message was "ignored").
Add _extract_trailing_user_content: on resume, collect the whole trailing
run of consecutive user messages (those after the last assistant/tool
message) and concatenate them (blank-line joined for text; merged content
blocks when any message is multimodal). Prior turns stay SDK-cached and
are not replayed.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(models): resolve supervisor wizard defaults
Replace the legacy multi-agent supervisor wizard's OpenAI and Databricks model pins with provider-catalog suggestions while preserving the free-form model prompt.
Unknown custom endpoints now receive no unrelated vendor default and require an explicit model. Add endpoint-specific coverage and remove both wizard entries from the hardcoded-model baseline.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(onboarding): require a supervisor model
Reject empty supervisor model input before generating an openai-agents spec. Custom endpoints must now provide an explicit model, and known providers fall back to operator input if their catalog has no default.
Keep the user on the model-selection step with a clear validation message and cover both custom-endpoint and empty-catalog retries.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* docs(onboarding): map supervisor provider branches
Document how the helper's profile, default OpenAI, and custom-endpoint states correspond to the wizard menu. This makes the explicit-input fallback clear when future endpoint choices are added.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
#2976 moved git untracked-cache setup off the runner startup path into a
daemon thread. The worker now shells out to git at an arbitrary moment, so
it can land inside a test that has swapped the process-global
subprocess.run and be recorded as one of that test's own calls.
That is how it failed CI on an unrelated PR: the databricks login test
asserts on the argv it captured and instead saw a stray
`config core.untrackedCache true`.
Stub GitFilesystemRegistry.start for the suite by default, with an
untracked_cache_start fixture for the worker's own tests, and harden the
login recorder so foreign argv reaches the real runner rather than the
capture list.
Signed-off-by: Ross Sclafani <rsclafani@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase 1 of the modular native-harness registry refactor landed (10 PRs,
2026-07-28 → 07-31). Bring the design doc in line with what actually shipped:
- Status header, Phase 1 subtotal, effort summary, and bottom line updated from
forward-looking ('1.1–1.3 in review') to Phase 1 complete / Phase 2 next.
- Ledger: 1.8 (#3648) landed; 1.4 marked descoped (with rationale); the 1.7
opencode-e2e follow-up (#3656) recorded; per-PR merge dates added.
- Calibration rewritten as a Phase 1 retrospective: estimate (~20–29 eng-days)
vs. actual (10 PRs / 4 calendar days), the real cost centers (test-shape churn
+ review-caught behavior bugs, enumerated per PR), the correct runner re-scope,
the two intentional behavior deltas (qwen label, antigravity relay), and the
recurring uv.lock / full-suite-only-flake operational friction.
Doc-only; no code change.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* feat(models): discover Cursor picker models from CLI
Replace the generated Cursor base-model catalog with live cursor-agent models discovery on the bound runner. Normalize compound effort variants and legacy dotted Claude spellings into the base-id namespace used by launch, /model switching, and terminal mirroring.
Route the discovered options through the existing session model-options cache, return retryable failures without blocking Cursor launch, and expose the same live listing to model-catalog callers. Remove the obsolete generator and eleven Cursor hardcode baseline entries.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): preserve valid Cursor picker options
Carry forward round-trip protection without restoring the stale generated denylist. Skip and log reversed Claude ids that the Cursor command parser cannot inject, while retaining CLI-advertised models verified against the current agent.
Keep cached picker options visible during asynchronous refreshes so model or effort changes do not transiently blank Cursor's picker. Cover both the parser guard and refresh behavior with regression tests.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): verify exact Cursor picker matches
Resolve the selected model's display name from the same live CLI catalog that supplies the Web picker, then refuse to press Enter unless Cursor highlights that exact row. This prevents fuzzy matching from silently selecting another model without restoring release-specific denylist entries.
Keep default/current tags unique in catalog order and limit stale-option retention to Cursor refreshes, preserving the existing drop-on-refresh behavior for Codex and other runner-backed pickers.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(models): centralize owned static fallbacks
Move the remaining Claude and Codex release-curated aliases into one fallback module. Record the responsible adapter, catalog provenance, and the concrete discovery gap for every registered fallback.
Carry that metadata through ModelListing and expose it in sys_list_models payloads, while preserving existing model ordering and provider behavior. Move—not expand—the nine lint baseline entries so future fallback edits remain explicit.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(cursor): reuse live model metadata when switching
Cache Cursor model display names when the runner serves the live picker catalog and pass the selected label into the TUI bridge. This avoids spawning cursor-agent models a second time for the same selection while retaining live discovery as a cold-cache fallback.
Also require the highlighted picker row to match the complete normalized display label or one of its suffixed variants, preventing similarly prefixed models from being accepted. Clear the cache with the existing session and agent lifecycle caches and cover cached, cold-cache, and fuzzy-match behavior.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(models): key fallbacks by provider constants
Use the canonical subscription and CLI-config kind constants as static fallback registry keys. This removes a silent coupling where renaming a provider kind could otherwise turn an owned fallback into an empty listing.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* lint(models): enforce owned fallback boundary
Allow unavoidable static model aliases only when AST analysis proves they are confined to complete StaticModelFallback records in the central model_fallbacks module. Require literal owner, provenance, and discovery-gap metadata, and reject fallback tuples reused outside those records.
Remove the nine centralized fallback rows from the count-based baseline while retaining the temporary baseline for independent migrations that have not landed yet. Add focused positive and bypass-resistance tests and document the structural exception.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* test(models): scan the owned fallback registry
Run the structural hardcode scanner against the production model_fallbacks module, proving the real stacked records satisfy the owned fallback boundary without count-based allowances.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* test(models): require the fallback registry
Make the production-registry lint assertion fail if model_fallbacks.py is missing instead of passing vacuously. Clarify that only module-level literal tuples qualify for the structural exemption so nested aliases intentionally fail closed.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* Update Codex fallback aliases
Replace the stale Codex fallback entries with the current GPT 5.6 Sol, Luna, and Terra aliases. Keep the change scoped to Codex; no Gemini fallback is introduced.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Creating a project against a container-deployed server failed with 405.
create_app mounts the projects router only when a project store is wired,
and the Docker entrypoint built every other store but never this one — so
POST /v1/projects was not a route at all and fell through to the SPA
catch-all (GET-only), which answers 405. The CLI server path already wires
it, so the same build worked under `omnigent server start` and failed in
the container.
Construct SqlAlchemyProjectStore from the resolved database URL and pass it
to create_app, mirroring the other stores.
Signed-off-by: apeltekci <andrew@peltekci.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
Final Phase-1 PR of the modular native-harness registry refactor: move
registry-parallel enumerations onto HarnessCapabilities.
- Add a fork_history axis (ForkHistory enum: none/rebuild/preamble) to
HarnessCapabilities, declared per harness in _BUILTIN_CAPABILITIES. Derive the
server's two fork-history gating frozensets in _sessions/common.py from it
instead of hand-listing. The derivation emits each canonical id plus its
reversed native-<key> spelling, because native-claude/native-codex/native-cursor
are valid ids canonicalize_harness passes through unchanged and the read sites
match on the canonicalized id (guarded by the existing reversed-spelling fork
test) — so the derived sets are a superset of the prior literals.
- Add optional shell_tool_name / shell_tool_prompt fields carrying the harness
bench's shell-tool provocation; delete the bench's hardcoded
_NATIVE_TOOL_PROVOCATION table and read the fields off capabilities in
native_vendor() (byte-identical (tool_name, prompt) per harness).
- Delete the dead _HARNESS_MODULES literal in runtime/harnesses/__init__.py
(~120 lines, overwritten unconditionally by harness_modules() next line).
- Extend the drift-guard tests in test_harness_capabilities.py.
Scope kept tight to the doc's mandate: sets that would need new NativeCodingAgent
identity fields (_ANTIGRAVITY_FAMILY_HARNESSES, _PROVIDER_RESOLUTION_HARNESS,
*_NATIVE_TERMINAL_ROLE) are left as-is; noted as follow-ups.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* feat(models): discover Cursor picker models from CLI
Replace the generated Cursor base-model catalog with live cursor-agent models discovery on the bound runner. Normalize compound effort variants and legacy dotted Claude spellings into the base-id namespace used by launch, /model switching, and terminal mirroring.
Route the discovered options through the existing session model-options cache, return retryable failures without blocking Cursor launch, and expose the same live listing to model-catalog callers. Remove the obsolete generator and eleven Cursor hardcode baseline entries.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): preserve valid Cursor picker options
Carry forward round-trip protection without restoring the stale generated denylist. Skip and log reversed Claude ids that the Cursor command parser cannot inject, while retaining CLI-advertised models verified against the current agent.
Keep cached picker options visible during asynchronous refreshes so model or effort changes do not transiently blank Cursor's picker. Cover both the parser guard and refresh behavior with regression tests.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): verify exact Cursor picker matches
Resolve the selected model's display name from the same live CLI catalog that supplies the Web picker, then refuse to press Enter unless Cursor highlights that exact row. This prevents fuzzy matching from silently selecting another model without restoring release-specific denylist entries.
Keep default/current tags unique in catalog order and limit stale-option retention to Cursor refreshes, preserving the existing drop-on-refresh behavior for Codex and other runner-backed pickers.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(models): centralize owned static fallbacks
Move the remaining Claude and Codex release-curated aliases into one fallback module. Record the responsible adapter, catalog provenance, and the concrete discovery gap for every registered fallback.
Carry that metadata through ModelListing and expose it in sys_list_models payloads, while preserving existing model ordering and provider behavior. Move—not expand—the nine lint baseline entries so future fallback edits remain explicit.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(cursor): reuse live model metadata when switching
Cache Cursor model display names when the runner serves the live picker catalog and pass the selected label into the TUI bridge. This avoids spawning cursor-agent models a second time for the same selection while retaining live discovery as a cold-cache fallback.
Also require the highlighted picker row to match the complete normalized display label or one of its suffixed variants, preventing similarly prefixed models from being accepted. Clear the cache with the existing session and agent lifecycle caches and cover cached, cold-cache, and fuzzy-match behavior.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(models): key fallbacks by provider constants
Use the canonical subscription and CLI-config kind constants as static fallback registry keys. This removes a silent coupling where renaming a provider kind could otherwise turn an owned fallback into an empty listing.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* Update Codex fallback aliases
Replace the stale Codex fallback entries with the current GPT 5.6 Sol, Luna, and Terra aliases. Keep the change scoped to Codex; no Gemini fallback is introduced.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Follow-up to #3599 (PR 1.7). That PR moved the built-in native agent-name
constants into a shared public block in omnigent/native_coding_agents.py and
migrated the claude/codex host e2e tests onto them, but missed the opencode
sibling: test_host_opencode_native_e2e.py still defined a local
_OPENCODE_NATIVE_AGENT_NAME = "opencode-native-ui" literal and asserted a stale
'_ensure_default_opencode_agent did not run' message (that per-harness seeder
was collapsed into _ensure_default_native_agents).
Import the shared OPENCODE_NATIVE_AGENT_NAME constant and update the message so
all three host e2e tests are consistent. Test-only; opt-in e2e (skipped without
OMNIGENT_E2E_OPENCODE_NATIVE=1).
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* refactor(harness): registry-driven server seeding loop (PR 1.7)
Collapse the server's built-in native-agent seeding onto the
NativeHarnessProvider seam. The 11 hand-written _ensure_default_<x>_agent
helpers + their 11 _build_<x>_native_bundle partners become two
registry-driven functions in omnigent/server/app.py:
- _build_native_bundle(provider): resolves provider.materialize_agent_spec via
the seam and runs the shared materialize -> bundle -> tar dance. The
per-harness `model` arg variance (codex required kw / kiro,opencode default /
the rest none) is bridged by one inspect.signature check.
- _ensure_default_native_agents(...): loops NATIVE_CODING_AGENTS, resolving the
provider by key and seeding each content-aware via _ensure_builtin_agent.
debby / polly / _ensure_extra_builtin_agents stay hand-written. Removed the now
-dead _<X>_NATIVE_AGENT_NAME constants and the *_NATIVE_CODING_AGENT imports.
Net server/app.py -455/+146.
Redeploy safety: builtin_agent_id(name) is a pure hash of the agent name, and
the names (NativeCodingAgent.agent_name) and bundle bytes are unchanged, so
seeded ids and bundles stay byte-identical (verified: sha256 of
_build_native_bundle output matches the pre-loop named builders across all
model-arg variants). New tests freeze the 11 expected ids and assert the loop
covers every native agent. Updated test_builtin_bundles / test_app to the
generic builder; fixed stale symbol refs in two e2e tests and a scheduled-tasks
integration test.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* test(server): cover the registry-driven native seeding paths (PR 1.7)
The seeding-loop collapse removed ~330 lines that were only exercised
transitively by e2e suites; add direct unit coverage so the new generic path is
fully covered and the coverage gate recovers:
- Parametrize the native bundle-builder tests over EVERY native agent (was a
4-agent sample), so each harness's _materialize_* + bundle path is covered
directly, across both model-arg shapes.
- Cover the two defensive guards in _build_native_bundle /
_ensure_default_native_agents (missing materialize hook, missing provider row).
- Add an end-to-end seed test asserting all 11 native agents register under
their stable builtin_agent_id with a retrievable bundle.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* docs(server): note the model-axis limit of the native seed signature bridge
Address Polly non-blocking note: the inspect.signature bridge in
_build_native_bundle understands only the `model` kwarg; a future harness
whose materializer needs a different required kwarg fails loud at seed time
rather than routing. Comment so the next author knows.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* refactor(harness): aggregate built-in native agent-name constants (PR 1.7)
The seeding-loop collapse deleted the 11 private _<X>_NATIVE_AGENT_NAME
constants from server/app.py (the loop uses agent.agent_name directly), which
pushed callers that need one specific built-in onto magic-string literals
("claude-native-ui", "qwen-native-ui", ...) in the tests.
Restore them as PUBLIC constants in omnigent/native_coding_agents.py — the
module that already indexes the registry rows — so seeding and tests share one
named, registry-derived source of truth instead of re-deriving the literal:
- Add CLAUDE_NATIVE_AGENT_NAME ... KIMI_NATIVE_AGENT_NAME (each = the row's
agent_name) to native_coding_agents.
- Point the server + scheduled-tasks tests at the shared constants (drop the
bare "qwen-native-ui" / "antigravity-native-ui" / "claude-native-ui" strings).
- Fold the two host e2e tests' own local _CLAUDE/_CODEX_NATIVE_AGENT_NAME
literals onto the shared constants too.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
Two localized optimizations to SqlAlchemyConversationStore.list_items,
which backs GET /v1/sessions/{id}/items (the web chat transcript read).
- Scope the after/before cursor subqueries to conversation_id so they
land on the (workspace_id, conversation_id, id) primary key as point
lookups. Without it, (workspace_id, id) leads no index and each
paginated page degraded to a workspace-wide scan.
- load_only the seven columns _to_item reads, dropping the wide
search_text Text column that this read path never touches. On
Postgres search_text is TOAST-ed, so omitting it skips a detoast and
roughly halves the bytes pulled per row on a chatty conversation.
Scoping the cursor to the conversation also fixes a latent correctness
edge: a cursor id from another conversation previously resolved its
position workspace-wide and applied it as a cutoff; it now yields an
empty page, guarded by a new test.
Co-authored-by: Isaac
* fix(cli-auth): write the session-JWT file privately instead of chmod-ing it after
_store_entry's docstring already promised the file is written "with user-only
read/write permissions (0o600) - the file may hold session JWTs, which are
sensitive". The implementation did not deliver that:
path.parent.mkdir(parents=True, exist_ok=True)
...
path.write_text(json.dumps(data, indent=2))
os.chmod(path, stat.S_IRUSR | stat.S_IWUSR)
write_text creates a missing file at the process umask, so on the very first
login - exactly when a session JWT is first persisted - the token sat on disk
readable by every local user until the chmod landed. Measured:
dir mode after mkdir : 0o755
file mode after write_text: 0o644 <- JWT is on disk at this mode
file mode after chmod : 0o600
The parent ~/.omnigent was also left world-traversable, and clear_token
rewrote the same file with no chmod of its own, relying on the mode of a file
it may not have created.
Routes both writers through _write_tokens_file, mirroring the pattern already
used in claude_native_bridge._atomic_write_user_json: a tempfile beside the
target (created owner-only by tempfile before any bytes are written), fsync,
chmod, then os.replace. The directory is created 0o700.
The rename also fixes a robustness bug: write_text truncated in place, so a
write that failed partway left a truncated file, and the JSONDecodeError
handler in _store_entry treats that as {} - silently discarding every stored
token for every server. The temp is discarded on failure and the previous file
is left intact.
Tests: tests/test_cli_auth_token_file_mode.py. Three of the eight fail on the
previous code (the on-disk window, the directory mode, and token loss on a
failed write); the rest pin the final mode, round-tripping, trailing-slash
normalisation and selective clearing.
Signed-off-by: Mr-Neutr0n <64578610+Mr-Neutr0n@users.noreply.github.com>
* style: satisfy ruff format
Pre-commit's ruff-format hook flagged the skipif decorator in the new test
module; it fits on one line.
Signed-off-by: Mr-Neutr0n <64578610+Mr-Neutr0n@users.noreply.github.com>
* refactor(cli-auth): hoist state-dir hardening into _write_tokens_file
Move the 0o700 mkdir + chmod from _store_entry into _write_tokens_file
so every writer routes through it. Previously only _store_entry
hardened the directory, so a clear_token-only interaction left a
pre-existing world-traversable (0o755) ~/.omnigent untightened. Adds a
regression test pinning that clear_token now hardens the dir.
Co-authored-by: Isaac
---------
Signed-off-by: Mr-Neutr0n <64578610+Mr-Neutr0n@users.noreply.github.com>
Co-authored-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* feat(models): discover Cursor picker models from CLI
Replace the generated Cursor base-model catalog with live cursor-agent models discovery on the bound runner. Normalize compound effort variants and legacy dotted Claude spellings into the base-id namespace used by launch, /model switching, and terminal mirroring.
Route the discovered options through the existing session model-options cache, return retryable failures without blocking Cursor launch, and expose the same live listing to model-catalog callers. Remove the obsolete generator and eleven Cursor hardcode baseline entries.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): preserve valid Cursor picker options
Carry forward round-trip protection without restoring the stale generated denylist. Skip and log reversed Claude ids that the Cursor command parser cannot inject, while retaining CLI-advertised models verified against the current agent.
Keep cached picker options visible during asynchronous refreshes so model or effort changes do not transiently blank Cursor's picker. Cover both the parser guard and refresh behavior with regression tests.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): verify exact Cursor picker matches
Resolve the selected model's display name from the same live CLI catalog that supplies the Web picker, then refuse to press Enter unless Cursor highlights that exact row. This prevents fuzzy matching from silently selecting another model without restoring release-specific denylist entries.
Keep default/current tags unique in catalog order and limit stale-option retention to Cursor refreshes, preserving the existing drop-on-refresh behavior for Codex and other runner-backed pickers.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(cursor): reuse live model metadata when switching
Cache Cursor model display names when the runner serves the live picker catalog and pass the selected label into the TUI bridge. This avoids spawning cursor-agent models a second time for the same selection while retaining live discovery as a cold-cache fallback.
Also require the highlighted picker row to match the complete normalized display label or one of its suffixed variants, preventing similarly prefixed models from being accepted. Clear the cache with the existing session and agent lifecycle caches and cover cached, cold-cache, and fuzzy-match behavior.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(cursor): drop unused parser binding
Keep the model-option setdefault call for deduplication without assigning its return value before the later result loop. This addresses the code-quality finding without changing parser behavior.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(cursor): handle missing CLI during model switch
Catch click.ClickException while refreshing a cold Cursor model catalog so a missing cursor-agent executable becomes the existing handled RuntimeError instead of escaping the runner endpoint as a 500.
Add bridge-level regression coverage for the preserved exception cause. The focused Cursor/native-event suite passes 122 tests and full pre-commit passes.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* feat(telemetry): emit PolicyRegisteredEvent and PolicyDeletedEvent
Add two new telemetry events that fire on policy create/delete for
both session-level and admin-level policies:
- PolicyRegisteredEvent: fired after a successful POST to
/v1/sessions/{id}/policies or /v1/policies. Records handler,
policy_type, scope ("session" or "admin"), session_id, and
anon_user_id so we can see which handlers are being registered and
at what scope.
- PolicyDeletedEvent: fired after a successful DELETE. Looks up the
existing policy first so the handler is available; silently skips
emission when the policy was already absent (idempotent delete).
Both events follow the existing try/except BLE001 fire-and-forget
pattern used by SessionStoppedEvent and SessionDeletedEvent.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor(policy-store): return deleted Policy from delete/delete_default
Previously delete() and delete_default() returned bool, causing a
second PK lookup in the route layer to retrieve the handler before
emitting telemetry. Changing the return type to Policy | None
eliminates that extra round-trip: the store already loads the row to
perform the delete, so we can return the entity at no additional cost.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor(telemetry): drop handler from PolicyDeletedEvent, revert store changes
handler required a pre-fetch before delete to avoid an extra DB
round-trip, which meant changing the store layer. Dropping the field
keeps PolicyDeletedEvent simple and the store interface unchanged.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
The host orphan reaper's waitpid fallback uses os.WNOHANG and
os.waitpid(-1, ...), neither of which exists/works on native Windows.
Windows also has no child reparenting to a subreaper, so there is
nothing to reap. The periodic sweep swallowed the resulting
AttributeError, but the final drain in run()'s finally block runs
unguarded and would crash shutdown.
Return early with 0 when os.WNOHANG is absent, matching the reaper's
own "non-Linux is a no-op" contract.
Co-authored-by: Isaac
* refactor(harness): DI-seam runner interrupt/stop — collapse 16 handlers (PR 1.6)
Route the runner's native interrupt / stop dispatch through a
dependency-injected NativeInterruptRunner instead of 16 per-harness closures
plus two hardcoded `if _harness == "<x>-native"` chains in the /events handler.
Mirrors the CodexGoalRunner DI precedent (omnigent/runner/codex/goal.py):
app-scope state (AP client, resource registry, event publisher, sub-agent wake
plumbing, codex bridge-state resolver) is injected at construction, typed via
Protocol.
- New omnigent/runner/native/interrupt.py: the 9 uniform interrupt and 7
uniform stop handlers collapse to two descriptor-driven methods
(_UNIFORM_INTERRUPT / _UNIFORM_STOP); claude interrupt (bridge-id) and codex
interrupt (MCP-startup + turn/interrupt) keep dedicated methods, moved
verbatim. interrupt()/stop() return None for handler-less harnesses so the
caller falls through to the in-process cancel.
- app.py: the two dispatch chains become one runner.interrupt()/.stop() call +
fall-through; the 16 closures are deleted (net app.py -470). Local
`from omnigent.<x>_native_bridge import` stays at call time so bridge-module
monkeypatches keep resolving (no test repoints).
- 12 new unit tests for NativeInterruptRunner.
- Doc: add 1.6 ledger row (gap-fill deferred); flip stale 1.5c row to landed.
Scope: migration-only, behavior-preserving. The antigravity/opencode coverage
gap (no interrupt/stop handler; they fall through to _cancel_inprocess_turn) is
left unchanged and pinned by a no-handler test; wiring agy interrupt_turn() /
opencode client.abort() is a deferred follow-up.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* docs(runner): fix uniform interrupt/stop harness counts in interrupt.py
Address Polly non-blocking doc nit: the module comments said 'nine uniform
interrupt' and 'seven uniform stop', but _UNIFORM_INTERRUPT has seven entries
and _UNIFORM_STOP six (claude/codex interrupt and claude stop are special-cased;
codex/pi alias stop to interrupt). Clarify uniform-vs-total counts. Doc-only.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* refactor(pi): route wire APIs from catalog metadata
Replace Pi's release-specific GPT Chat Completions allowlist with normalized Unity Catalog model-service wire metadata shared by native and inner Pi execution.
Thread generic-provider wire configuration through the harness, resolve dedicated AI Gateway URLs back to their workspace API origin, and avoid probing non-Databricks providers. When discovery is unavailable, route unknown GPT models to Responses while retaining the documented system-model compatibility fallback.
Cover Chat, Responses, dedicated-gateway, generic-provider, alias, outage-cache, and Responses-only catalog behavior. Verified 275 focused Pi/catalog tests, isolated runtime spawn-env tests, live production UC metadata, and repository-wide pre-commit.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(pi): hoist model routing imports
Move the catalog, gateway, subprocess, and compatibility imports used by Pi routing to module scope so dependencies are explicit and consistently initialized.
Extract the shared Pi model compatibility predicates into a small leaf module to avoid introducing a model_catalog/pi_native_credentials import cycle. Update tests to patch the module-bound credential resolver.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* Add agy (Google Antigravity CLI) as a 7th polly sub-agent
polly's roster now includes agy alongside claude_code, codex, opencode,
cursor, hermes, and pi. agy drives the antigravity-native harness
(Gemini-native, own Google account auth via ~/.gemini; does not run
Claude/GPT-family models) and follows the same
IMPLEMENT/REVIEW/EXPLORE contract as the other worktree-scoped
implementers, with gate_pushes: false so it can open its own PRs.
Updates the roster count, preflight check, trigger phrases, and
cross-vendor review/cancellation lists in config.yaml; the
investigate/fanout/cross-review skills' vendor lists; and the
structural e2e test assertions (roster tuple, harness family map,
policy-argument count) to match.
Signed-off-by: Imraul Emmaka <ikemmaka@ualr.edu>
* fix(antigravity-native): re-deliver turns agy rejects while verifying the account
agy's TUI composer mounts ~3s after launch, but its account-eligibility
check is not settled until ~7-9s. A turn submitted inside that window is
consumed by agy — the draft leaves the composer, so the submit verifies —
and answered with "We're finishing verifying your account eligibility"
instead of starting a cascade. Nothing retried, so the turn was silently
lost and the terminal sat idle.
Detect the notice after a submit and re-deliver until agy takes the turn,
bounded by 90s. The running-turn marker is checked first so a notice still
rendered from a prior attempt can never re-send a turn that already landed,
and the probe fails open so a future agy that renames its running footer
keeps delivering rather than retrying.
Programmatic first turns — a polly sub-agent dispatch — land in that window
on every launch; interactive users usually type slowly enough to miss it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Signed-off-by: Khoi Nguyen <khoifish@goodstorytime.com>
* fix(antigravity-native): treat agy's collapsed-paste placeholder as a rendered draft
agy replaces a paste carrying many line breaks with a single
`[Pasted text #N +M lines]` row instead of echoing the text into the
composer. The threshold is line-count based (~13+ line breaks); total
length does not matter, so a long single-line message still renders
verbatim while a multi-line one never does.
The render gate looks for the message's needle in the composer, which a
collapsed paste can never contain, so delivery raised "agy did not render
the pasted message in its input box before submit" while the draft was in
fact sitting there. Sub-agent task prompts are exactly this shape, so a
polly dispatch failed on its first turn every time; the single-line
follow-up prompts it sent next happened to render verbatim and worked,
which made it look like a startup race.
Recognise the placeholder as draft content in _draft_in_input_region so
both the render gate and the submit verification key off it appearing and
then leaving the composer — the submit stays verified rather than blind.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Signed-off-by: Khoi Nguyen <khoifish@goodstorytime.com>
* fix(antigravity-native): bind the TUI injector to an explicit bridge dir
The interaction bridge's default TUI injector resolved the bridge directory
from HARNESS_ANTIGRAVITY_NATIVE_BRIDGE_DIR, on the assumption (stated in its
docstring) that "the reader/CLI both run with it set". That is stale: the
reader now runs as a task INSIDE the runner process, which never carries that
variable — it is set only for the harness subprocess by
build_antigravity_native_spawn_env.
So every web approval failed with "HARNESS_ANTIGRAVITY_NATIVE_BRIDGE_DIR is
required" — 100% of the time, not intermittently. The RPC delivery flipped
agy's backend step, but agy's own permission prompt was never dismissed, so
the terminal did not advance and the next typed turn risked landing in the
stale prompt's buffer.
Add tui_injector_for(bridge_dir) and have the reader — which is handed its
bridge_dir — use it. _inject_via_tui stays for callers that genuinely run
with the harness env, with its constraint now spelled out.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Signed-off-by: Khoi Nguyen <khoifish@goodstorytime.com>
* fix(antigravity-native): close a turn agy reports finished (quiescence backstop)
Turn completion was inferred purely by pattern-matching step types.
_is_turn_close_step has already accreted three special cases — clean text
close, ERROR planner, degenerate DONE — and its own docstring explains that
missing one leaves turn_active stuck True forever: the spinner never clears
and the NEXT turn cannot re-open RUNNING either. Every agy step type it does
not know about is a permanently stranded session, and that list only grows.
agy already publishes the answer. Every GetAllCascadeTrajectories summary
carries a per-cascade CASCADE_RUN_STATUS, which appeared in this codebase
exactly once — in a docstring example — and was never read, even though the
rotation detector already fetches those summaries on every scan.
Use it as a BACKSTOP: when agy reports the bound cascade idle on two
consecutive scans while Omnigent still believes a turn is open, close it. The
step-based close stays the fast path; this only catches what it missed. Being
reconciliation rather than edge detection, it is idempotent and self-healing —
a missed, unknown, or reordered step now costs one detector interval instead
of stranding the session.
Verified against agy 1.1.8 that the status reports RUNNING both while working
and for the entire time a permission gate is parked (75s observed), so the
backstop cannot close a turn that is waiting on a human. Two consecutive ticks
are required so the gap between delivering a turn and agy starting it is not
mistaken for the end of one.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Signed-off-by: Khoi Nguyen <khoifish@goodstorytime.com>
* fix(antigravity-native): avoid duplicate verification retries
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Imraul Emmaka <ikemmaka@ualr.edu>
Signed-off-by: Khoi Nguyen <khoifish@goodstorytime.com>
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Co-authored-by: Imraul Emmaka <ikemmaka@ualr.edu>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* feat(web): add zoom controls to subagent graph panel
Add zoom in/out and fit-to-view buttons to the subagent graph panel
using ReactFlow's useReactFlow hook. Widen the zoom range from
0.3–1.5x to 0.1–3x so users can zoom in closer to read small nodes
or zoom out further for large graphs.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* style(web): fix prettier formatting for zoom control buttons
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
---------
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(openshell): pass workspace to SandboxClient lifecycle methods
The openshell SDK >=0.0.86 added a required `workspace` keyword argument
to `SandboxClient.create()`, `get()`, `delete()`, and `wait_ready()`.
Omnigent never passed it, so `sandbox create --provider openshell`
crashed with `TypeError: SandboxClient.create() missing 1 required
keyword-only argument: 'workspace'`.
Thread a workspace through _OpenShellClient and OpenShellSandboxLauncher,
resolved from: explicit constructor arg (YAML `sandbox.openshell.workspace`),
then `$OMNIGENT_OPENSHELL_WORKSPACE` env var, then "default".
Fixes#3513
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix: bump openshell floor to >=0.0.88 and close test gaps
The `workspace` kwarg landed in openshell 0.0.88, not 0.0.86 — 0.0.86
still has the old signature and would crash with `got an unexpected
keyword argument 'workspace'`. Bump the floor accordingly.
Also record the workspace reaching the fake SDK and assert it in both
the _OpenShellClient and managed_hosts tests.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* chore: strip index-dependent size fields from uv.lock
pypi.org's index serves wheel/sdist sizes while proxy indexes may not,
so re-locks were flipping ~2,900 'size = N' lines back and forth. The
sizeless form is canonical on main; this keeps the diff to the real
dependency changes.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Co-authored-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(web): keep the session config gear usable while the session is asleep
The gear required liveness === "online", so an asleep session couldn't
change model/effort even though PATCH /v1/sessions persists overrides
and the next wake applies them. Gate the gear like the composer (inert
only for read-only viewers and unreachable sessions) and make the
native model catalog survive runner death so the picker stays filled:
- relay exit / refresh_state with no runner now mark the per-session
catalog stale instead of deleting it; snapshots keep serving it
- a stale catalog is re-fetched in the background once a live runner
is bound again, and replaced on success
- an asleep claude-native session with a cold cache (server restart)
refills from its host over the host tunnel - the same pre-launch
source the new-session picker uses
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(e2e): scope the mermaid preview assertion to the diagram svg
The rendered Streamdown mermaid block carries chrome icon svgs (zoom /
copy controls) next to the diagram, so the strict single-svg locator
fails with "resolved to 3 elements" on every run since #3498 merged.
Target the diagram svg via mermaid's aria-roledescription stamp, which
also makes the assertion check the diagram itself rather than any svg.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
Follow-up to the non-blocking review notes on #3479.
- codex_executor consumes agent_env.declared_passthrough instead of keeping
its own copy. It already imports agent_env, so the reason the duplicate
existed no longer applies. Test repointed at the shared helper.
- POLICIES.md now explains that agent CLIs get a deny-by-default environment
and what env_passthrough is for. The migration note only ever lived in a PR
description, so the two cases that bite -- a generic ACP agent with no vendor
family, and a goose authenticated by an ambient provider key rather than
gateway routing -- were undocumented.
Signed-off-by: Mr-Neutr0n <harikp2002@gmail.com>
pypi.org's simple index serves a size for every file while proxy
indexes may not, so each re-lock added or stripped 'size = N' across
~2,900 lines depending on which index resolved it. Make the sizeless
form canonical (the hash is the integrity check): the fixer now drops
size fields and --check flags them, so re-locks from either side
converge on one form.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
ReactFlow's pan-on-drag behavior was intercepting pointer events on
graph nodes, preventing the existing <Link> wrapper from navigating.
Adding the `nopan nodrag` utility classes tells ReactFlow to leave
those events alone so clicks reach the router link.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
Mirror the pvc_mounts config knob for Kubernetes Secrets: project a
pre-created Secret as a read-only file volume on the runner's host
container. A Secret volume (no subPath) is refreshed in place by the
kubelet, so a long-lived runner picks up a rotated credential without a
restart — unlike envFrom, which is frozen at container start.
- server: parse/validate sandbox.kubernetes.secret_mounts at config load
(DNS-1123 name, absolute/normalized/non-reserved path, intra-list and
pvc<->secret path-collision checks), failing loud at startup
- onboarding: add the secret volume + host-container-only volumeMount in
build_pod_manifest (optional=False, defaultMode 0440), threaded through
the launcher
- tests mirror the pvc_mounts coverage
Signed-off-by: bdchatham <bdchatham@gmail.com>
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
Each open session in the web UI holds a long-lived event-stream HTTP
response. Over HTTP/1.1 browsers cap concurrent connections at ~6 per
origin, so opening several windows/tabs against a raw :8000 deploy fills
the pool with held-open streams and every other request stalls — the UI
appears frozen across all windows while the server is idle.
The bundled Caddy overlay and every managed platform already terminate
TLS with HTTP/2, which multiplexes the streams and dissolves the cap;
the gap was only that nothing told operators this proxy is also the fix.
Document it in the deploy README ("Serving") and point to it from the
Caddyfile. Docs-only; no server behavior change.
Co-authored-by: Isaac
* feat(sandbox): make recursive dotfile hiding opt-in, add mask_paths
The sandbox hid every dotfile under the working directory by walking the
whole tree. On medium-to-large projects that walk is slow and routinely
trips the entry cap, and it masks far more than the secrets it targets.
Make the recursive scan opt-in and add a way to hide specific paths:
- cwd_hidden_scan_recursive (default false) scans only the top level of
the cwd and each read_paths root (including $HOME when it is a granted
read path). The top-level dotfiles that hold most secrets (.git, .env,
.aws, .ssh, ...) are still masked, but the walker no longer descends the
whole tree. Set it true for untrusted trees where a deeply nested
credential file would be an unacceptable leak.
- mask_paths hides a named file or folder regardless of a leading dot,
resolved like read_paths (~ expanded, relative to cwd, no $VAR). Files
are masked as an empty file, folders as an empty view, on top of the
dotfile mask in every mode.
Both backends enforce the new fields: linux_bwrap binds /dev/null for
files and a tmpfs for folders; darwin_seatbelt emits literal/subpath deny
rules. Behavior change: with the non-recursive default, dotfiles nested
below the first level are now readable unless recursion is turned on.
Signed-off-by: Thomas Garnier <6202935+mxatone@users.noreply.github.com>
* docs(sandbox): note is_dir symlink behavior for mask_paths
Clarify that the explicit mask_paths classification uses is_dir(), which
follows symlinks — unlike the dotfile walker's follow_symlinks=False — and
that seatbelt emits a harmless literal deny for a missing entry where bwrap
drops it on the re-stat.
Signed-off-by: Thomas Garnier <6202935+mxatone@users.noreply.github.com>
---------
Signed-off-by: Thomas Garnier <6202935+mxatone@users.noreply.github.com>
* perf(web): render conversations before the full history window loads
Opening /c/<id> blocked first paint on fetchInitialHistoryWindow, which
pages backward (up to MAX_INITIAL_PAGES serial round-trips) until the last
two user prompts are on screen. On a real deployment each page is ~1s, so a
long tool-heavy last turn could stall the transcript for several seconds.
Fetch only the first page in the blocking bind, render immediately, then
page the rest of the window in the background behind a top-of-history
spinner. The previous-prompt heuristic is unchanged — just no longer on the
critical path.
- Extract the window-complete boundary into initialWindowComplete() and
reuse it in both fetchInitialHistoryWindow and the new backfill.
- bindStream fetches one page; backfillInitialWindow continues the same
paging loop after commit, holding loadingMoreHistory so scroll-up/rail
loaders don't double-fetch, generation-guarded like loadMoreHistory.
- New loadingInitialWindow flag drives a "Loading earlier messages…"
spinner above the oldest bubble.
Co-authored-by: Isaac
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* ⚡ perf(web): Unify initial history loading
- Build the prompt-boundary and viewport-fill window through one post-render loader
- Make the turn rail lazy and remove its eager 200-item history fetch
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* ✅ test(web): Cover lazy history loading
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* perf(web): pin the latest turn to the top with a trailing spacer
Add a LatestTurnSpacer as the last child of the message flow that pins the
newest turn's anchor to the top of the viewport (the newest real user prompt,
or the newest assistant text output when a page deep in a tool chain has no
prompt yet), letting the reply grow below it — the ChatGPT/Claude "question at
top" feel.
As a side effect the spacer keeps the transcript taller than its scroll
container whenever content sits above the anchor, so older history stays
reachable by scroll-up. That makes HistoryAutoLoader's viewport-fill fetch loop
redundant: it now pages only to the previous-prompt boundary (still capped by
initialWindowComplete), and the resize-driven re-fill and spinner-height
measurement are removed.
Spacer height = clientHeight − (anchor→content-bottom) − top gap, clamped to
≥ 0: it shrinks as the reply streams (its own top is fixed by the content
above, not by its height, so scrollHeight stays constant and stick-to-bottom
keeps the anchor pinned) and collapses to 0 once the reply exceeds the
viewport, restoring normal bottom-following.
Co-authored-by: Isaac
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* test(e2e-ui): regenerate visual baselines
* fix(web): keep loading history near the top
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* test(e2e-ui): regenerate visual baselines
* fix(web): preload history sooner near the top
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* refactor(web): show history skeleton for every page
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* fix(web): stabilize scroll during history prepends
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* style(web): loosen history skeleton spacing
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* style(web): use compact history loading indicator
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* fix(web): avoid latest turn spacer flicker
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* fix(web): observe initial history scroll adjustment
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* fix(web): bind history loading to live scroller
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
* 🐛 fix(web): freeze spacer to loaded turn
---------
Signed-off-by: Daniel Lok <daniel.lok@databricks.com>
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
Replace the stale exact Qwen context-window registry with metadata from the shared MLflow provider catalog. Keep only the self-describing Anthropic [1m] marker and the conservative 128K offline fallback.
Reuse the onboarding catalog cache for both context sizing and pricing, preserve cache pricing fields in ModelInfo, and support provider-qualified ids, OpenRouter vendor namespaces, and Databricks aliases without release-specific model mappings.
Ratchet the hardcoded-model baseline and document the migration behavior. Cover exact, family, namespace, ambiguity, cache, encoded-metadata, and offline resolution paths.
Tests: 59 focused provider/context-window tests; 110 model-catalog, compaction, and session-override tests; changed-file pre-commit; repository-wide pre-commit except the pre-existing stale routing_pb2.py binding; live MLflow lookup smoke test.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Collapse the terminal-ensure / attach path in create_session_terminal —
11 hardcoded `if terminal_name == "<x>" and session_key == "main"` arms —
behind a single generic `_ensure_native_terminal(...)` shell dispatched
through the NativeHarnessProvider seam. The attach-path sibling of the 1.5b
launch shell (#3500/#3501); reuses the `_launch_<x>` adapters and
NativeLaunchContext. codex/antigravity supply an ownership predicate; codex
supplies a `finalize` for its one-shot policy notice — both run under the
per-session ensure lock, matching the inline arms.
- New shell in runner/native/orchestration.py (view-based existence check,
returns JSONResponse: 200 / 500 / 409), exported from runner/native.
- app.py: 11 arms (~450 lines) -> one collect-then-dispatch block.
- Repoint the HTTP attach-path claude/codex auto_create monkeypatch targets
to the orchestration module (the seam resolves the adapter there).
- 8 new unit tests for the shell.
- Doc: add 1.5c ledger row; flip stale 1.5b-i/ii rows to landed.
Behavior-preserving: qwen error label -> "Qwen Code" (display_name, as 1.5b-i);
antigravity now wires ensure_comment_relay via the base ctx (the landed
_launch_antigravity adapter already passed it).
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
Normalize Databricks Unity Catalog supported_api_types into the provider-neutral ModelWireAPI vocabulary and retain those facts while converting runner catalogs into the id-only routing-client shape.
Replace the exact Pi model exclusion table with a catalog-backed Claude wire check. Pi now keeps Responses-capable GPT models on its supported Responses path, while endpoints explicitly lacking Anthropic Messages are redirected to claude-sdk. Missing metadata from older runners remains unknown and does not trigger a redirect.
Ratchet six retired hardcode allowances and update the migration plan.
Tests: 104 catalog and smart-routing tests; 7 Pi Responses/provider tests; changed-file pre-commit suite. The repository-wide pre-commit run passed every relevant hook and only reported the pre-existing stale routing_pb2.py baseline.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Remove the release-specific model from the Kimi launcher example so an unoverridden session uses the default already configured in the Kimi CLI.
Document the ownership boundary, assert that the spawn environment omits HARNESS_KIMI_MODEL when no model is declared, and ratchet the retired lint allowance.
Tests: 12 Kimi spawn-environment tests; structural example load; staged pre-commit including YAML and hardcoded-model checks.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* feat(models): discover ad-hoc CLI default
Resolve the model for minimal harness-less agent YAMLs from the Databricks OpenAI-family catalog during bundle materialization instead of baking a release-specific endpoint into the CLI.
Preserve --model and OMNIGENT_MODEL precedence, and fail with explicit configuration guidance when discovery is unavailable. Ratchet the removed pin from the hardcoded-model allowlist and document the new behavior.
Tests: 111 discovery-disabled CLI tests; staged pre-commit including the hardcoded-model lint.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* test(models): pin YAML model precedence
Add a direct regression test showing that an executor.model declared in agent YAML remains authoritative over OMNIGENT_MODEL and catalog discovery.
Fail the test if catalog resolution runs, so future changes cannot silently turn an explicit YAML model into a fallback lookup.
Tests: 114 CLI chat tests; targeted pre-commit checks.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* feat(models): discover onboarding defaults
Select provider setup defaults from the live catalog after filtering specialty modalities, using stable family preferences for broadly accessible Anthropic and OpenRouter choices instead of release-specific model pins.
When discovery is unavailable, leave onboarding unpinned so the user supplies an explicit model. Add deterministic catalog fixtures for interactive CLI coverage, ratchet three lint allowances, and document the migration.
Tests: 147 onboarding and configure-models tests; staged pre-commit including the hardcoded-model lint.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): require intended OpenRouter family
Keep OpenRouter onboarding defaults within the catalog's Kimi family. If discovery returns no compatible family member, require the user to enter a gateway model instead of silently selecting a newer proprietary entry.
Correct the setup comments to match Click's prompt behavior: blank input accepts a discovered default, while an unavailable default requires an explicit value.
Tests: 86 provider and resolver tests passed. Targeted pre-commit passed for all modified files.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* test(models): pin offline runtime failure
Cover Anthropic and OpenAI runtime fallback when neither the agent nor provider config names a model and catalog discovery returns no data. Both paths must fail closed with guidance to configure an explicit model or retry discovery.
Document that removing source pins affects shared runtime defaults in addition to onboarding prompts, and clarify that required-family policy tokens use case-insensitive substring matching.
Tests: 88 focused runtime, provider, and resolver tests passed. A broader 153-test run reached 152 passes plus one unrelated host-credential leak in the existing Claude fallback test. Targeted pre-commit passed.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* test: make sandbox-cwd assertions portable across macOS firmlinks
_resolve_sandbox_cwd ends in Path.resolve(), and macOS routes the test's
literal paths through firmlinks (/home via the automounter, /tmp ->
/private/tmp), so the literal-string assertions fail on any macOS dev
box while Linux CI stays green. Compare against the same resolution
instead; on Linux both sides are identical strings.
Signed-off-by: webdevtodayjason <jason@webdevtoday.com>
* test: tidy sandbox cwd portability assertions
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: webdevtodayjason <jason@webdevtoday.com>
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Co-authored-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* feat(models): discover ad-hoc CLI default
Resolve the model for minimal harness-less agent YAMLs from the Databricks OpenAI-family catalog during bundle materialization instead of baking a release-specific endpoint into the CLI.
Preserve --model and OMNIGENT_MODEL precedence, and fail with explicit configuration guidance when discovery is unavailable. Ratchet the removed pin from the hardcoded-model allowlist and document the new behavior.
Tests: 111 discovery-disabled CLI tests; staged pre-commit including the hardcoded-model lint.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* test(models): pin YAML model precedence
Add a direct regression test showing that an executor.model declared in agent YAML remains authoritative over OMNIGENT_MODEL and catalog discovery.
Fail the test if catalog resolution runs, so future changes cannot silently turn an explicit YAML model into a fallback lookup.
Tests: 114 CLI chat tests; targeted pre-commit checks.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* test(e2e): pin sessions mock model
Give the sessions-default REPL fixture an explicit mock-server model so the test exercises session routing rather than ad-hoc model discovery.\n\nThe E2E workflow intentionally disables catalog lookup. After ad-hoc defaults moved to catalog resolution, the model-less fixture exited before the REPL opened. Other approval fixtures in this file already pin the same mock-compatible model.\n\nTest: OMNIGENT_DISABLE_CATALOG_LOOKUP=1 OMNIGENT_SKIP_WEB_UI=true uv run --frozen pytest -q tests/e2e/test_repl_sessions_approval_e2e.py::test_sessions_default_flag_works --tb=short\nTest: pre-commit run --files tests/e2e/test_repl_sessions_approval_e2e.py
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Second half of the runner launch seam, completing 1.5b. Routes the 3 special
arms and the turn-path opencode cold-boot through the seam, and consolidates all
11 create-session legs into one dispatch. Behavior-preserving.
- orchestration: extend the shell _launch_native_terminal with pre_launch (an
async (has_terminal) -> PreLaunchResult callback run inside the lock, so the
has_terminal-dependent rebuild/transfer/needs checks see the same state the
inline arms did), build_context (lazy full-context enrichment for claude's
bundle_dir/agent_name/skills + closures and codex's bundle, run only on
create), and reraise (turn-path opencode converts a launch failure to a 503
instead of publishing a start-error event).
- app.py: replace the 11 per-harness create-session legs with a single
collect-then-dispatch block — each leg only assigns its lock dict, context,
and optional pre_launch/build_context/resolve_agent_spec, then one
_launch_native_terminal call runs them. The 3 special arms (claude rebuild+
transfer, codex needs-check, antigravity payload+transfer) supply their
has_terminal-gated pre_launch; claude/codex supply build_context (codex keeps
the outer spec_entry as agent_spec). Turn-path opencode uses reraise=True.
- Preserve terminal_ready: only claude populated it in the create-session
response, so only claude's dispatch result is captured back (the consolidation
fixes a regression where 1.5b-ii's first cut dropped it).
- Tests: repoint the app-level _auto_create_<x>_terminal monkeypatches that now
route through the seam — claude create-session (events_lifecycle 603/688,
session_resources 2198) and the create-session auto-create guard tests
(terminals_autocreate: claude + antigravity) — to the orchestration symbol the
adapter calls. Add shell unit coverage for build_context (enrich-only-on-create)
and reraise. The terminal-attach/route patches (1.5c path) are untouched.
Net app.py reduction continues; the 11-arm launch chain is gone. Pre-existing
codex gateway-env failures in events_lifecycle are unchanged (codex arm behavior
preserved; those tests are unrelated app-server/gateway artifacts).
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
The seven native resolvers (pi, hermes, kimi, cursor, goose, kiro, qwen) looked up their CLI with a bare shutil.which, while readiness and the SDK executors resolve through resolve_cli_binary's fallback ladder (the nvm/npm/homebrew bin dirs the daemon's frozen PATH omits). A CLI installed only in a ladder dir passes the readiness badge but fails at launch. Route the resolvers through resolve_cli_binary so the badge and the launch agree.
resolve_cli_binary gains a `which` hook so the resolvers keep their existing test seam; the fallback ladder always uses the real filesystem.
Signed-off-by: Andrew Demczuk <andrew.demczuk@gmail.com>
* fix(inner): stop agent-CLI subprocesses inheriting unrelated host secrets
Closes#3445.
pi and codex filtered os.environ before spawning their vendor CLI; goose,
kimi, qwen, acp and hermes did not, so every host secret - cloud tokens, other
providers' API keys - reached those processes, sandboxed or not. hermes was
worst: the no-HERMES_HOME branch passed env=None, which inherits everything.
Implements the decision on the issue.
agent_env.clean_agent_env(allow_prefixes, allow_exact, deny_exact,
extra_allowed, source)
The model is not "no credentials ever". It is a shared safe base (HOME, PATH,
proxy, locale, tmp, XDG, the omnigent-session marker), plus the harness's own
config/provider families, plus whatever the spec declared in
os_env.sandbox.env_passthrough.
Per-harness families, matching the table on the issue:
qwen QWEN_, OPENAI_, DASHSCOPE_
goose GOOSE_
kimi KIMI_, MOONSHOT_ (keeps its documented ambient auth)
acp none - base + env_passthrough only, the agent is arbitrary
hermes HERMES_ (see below)
pi and codex become thin calls. Their sets are preserved exactly, including
codex's OPENAI_API_KEY deny; verified by diffing the new output against the
original inlined logic over a synthetic environment - identical, with and
without passthrough. USER/LOGNAME/SHELL/TZ stay per-harness rather than
entering the shared base, because pi passes them and codex does not and this
refactor must not widen codex's set.
hermes prefix family: HERMES_ only, and deliberately not DATABRICKS_. Hermes
authenticates from files, not the environment - hermes_native_bridge copies
~/.hermes/auth.json and ~/.hermes/.env into the per-session HERMES_HOME
(hermes_native_bridge.py:386-394). HOME still passes, so nothing breaks, and
the credential family this change exists to contain stays contained.
Also restores the launcher's env-prune defense: the sandboxed paths bake
tuple(env.keys()) into with_spawn_env_allowlist, so a full-environ env made
that allowlist a no-op.
Tests: tests/test_agent_spawn_env_canary.py - parametrized over all seven
harnesses, planting nine credential-family canaries and asserting none
survive, plus that each still gets a usable environment, that a harness sees
its own family and not a sibling's, that kimi keeps ambient KIMI_/MOONSHOT_,
that env_passthrough works as the migration path, and that deny_exact beats a
matching prefix. 21 cases.
Executor suites: 967 passed, 13 skipped, 0 failed.
Signed-off-by: Mr-Neutr0n <64578610+Mr-Neutr0n@users.noreply.github.com>
* fix(inner): point the spawn-env canary at the real executors
Addresses review on #3479.
- Extract _build_spawn_env() on qwen/goose/acp/hermes, matching kimi's
existing shape, and parametrize the canary over the real builders with
secrets planted in a monkeypatched environ. The prefix table was a hand
copy, so a harness reverting to os.environ.copy() kept the suite green;
it now fails, which is what the module docstring already claimed.
- Add NODE_EXTRA_CA_CERTS to BASE_ALLOW_EXACT. Node honours it where
SSL_CERT_FILE is ignored, so without it a corporate-CA user upgrading
loses TLS on every Node harness without a NODE_ family of its own.
- Warn in acp_executor._ensure_initialized when the handshake fails or the
child dies first, naming os_env.sandbox.env_passthrough as the likely fix.
Signed-off-by: Mr-Neutr0n <harikp2002@gmail.com>
---------
Signed-off-by: Mr-Neutr0n <64578610+Mr-Neutr0n@users.noreply.github.com>
Signed-off-by: Mr-Neutr0n <harikp2002@gmail.com>
The policy evaluate endpoint is a BLOCKING hook: a harness waits on its
allow/deny before running a tool. Its payload rules were applied from a chain
of conditionals, and a rule in a branch only ever reaches whichever phase
lands in that branch. Three rules now come from one per-phase schema, and all
three run for every phase.
What was getting through:
- `event.data` was accepted as an object, a string or absent, then normalized
with `or {}`. For a tool phase that means the gate evaluated as though the
caller had sent nothing: every tool-name-scoped policy skipped, and the hook
answering allow. Tool and LLM phases now require an object; a bare string is
a legitimate wire form only on the prompt phase, and an absent payload is
malformed everywhere, since every first-party producer sends one.
- A tool-scoped gate needs a tool name, and only one spelling was accepted.
Producers differ: claude-native and the in-process tool dispatch send
`request_data.name`, the OpenCode plugin sends the tool in `event.target`.
Requiring the first rejected the second with a 400 — and that plugin turns
any non-2xx into ALLOW, so a stricter guard silently disabled every OpenCode
TOOL_RESULT policy rather than tightening it. Any declared source now
satisfies the rule, and the resolved name is written onto the container the
engine reads, so those policies gate instead of merely passing validation.
- `event.context` must be an object when present. An earlier revision of this
message described only two rules while the diff carried three.
`event.type` is also checked before being used as a dict key: an unhashable
value raised inside the lookup and surfaced as a 500 rather than a 400.
The three rules above were previously three independent structures (which
wire types are accepted; which phases need an object payload; where a tool
name may come from), each keyed by phase and each read with a permissive
`.get(phase, default)` fallback. The comment on them already said "one schema
per phase" — the code didn't enforce it: a phase added to the first structure
alone was silently accepted, validated as loosely as possible, and given no
tool-name rule at all, because the other two structures simply had no entry
for it and their lookups defaulted rather than erred. They're now one
NamedTuple per wire type with no default values on any field, so a new entry
cannot be added without deciding both properties at once, and the only
`.get()` left is the outer wire-type lookup, which 400s on a miss instead of
falling back to anything.
The test table enumerates each phase and non-object-data vector and
cross-multiplies them, rather than hand-listing every case — kept in sync
with the production schema by hand, since that schema lives inside a
route-registration closure and isn't something a test module can import. It
asserts the structured error code rather than the status alone, and now
includes a non-empty list alongside the empty one: both are simply
non-dict, but hand-listing only the empty list is coincidentally falsy in a
way a narrower, wrong fix (special-casing falsy values) would have passed.
Five mutations kill it: accepting object-or-string-or-absent everywhere,
requiring a single tool-name spelling, dropping the context rule, validating
the alternative spelling without normalizing it (caught because the oracle
asserts a tool-scoped DENY, not a 200), and giving one phase's schema entry a
wrongly permissive `data_must_be_object`.
A pre-existing test's docstring also claimed OpenCode's plugin sends REQUEST
data as a bare string; it now sends `{"text": ...}` like every other
first-party producer. Reworded to describe why the bare-string form is still
accepted (older/third-party compatibility) without attributing it to
OpenCode's current behaviour.
Signed-off-by: Andrew Reid <andrew@reid.ee>
Unconditionally set CLAUDE_CODE_USE_GATEWAY=1 in the Databricks ucode
subprocess env and stop setting CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS on
that path. Gateway-aware mode keeps tool search on so MCP schemas load on
demand, so the betas-disable knob is no longer needed here.
Update test_ucode_config_for_profile_reads_allowlisted_claude_state to
expect CLAUDE_CODE_USE_GATEWAY=1 in the ucode env instead of the removed
CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS flag.
Co-authored-by: Isaac
Signed-off-by: Bryan Qiu <bryan.qiu@databricks.com>
Co-authored-by: Bryan Qiu <bryan.qiu@databricks.com>
Co-authored-by: Harry Yao <harry.yao@databricks.com>
* fix(tunnel): give host/runner websocket tunnels a verifying SSL context
On interpreters whose OpenSSL default cert path is uninitialized (python.org
macOS framework builds before Install Certificates.command, and
python-build-standalone interpreters used by uv), ssl.create_default_context()
loads zero trust roots, so the host and runner wss:// tunnels failed with
CERTIFICATE_VERIFY_FAILED and looped on reconnect.
Add omnigent/tls.py (resolve_ca_file + cached client_ssl_context) that resolves
a CA bundle OS-trust-store-first with a certifi fallback, and pass that context
to both tunnel websockets.connect calls for wss:// (ws:// stays ssl=None).
egress/ca.py:_system_ca_bundle now shares resolve_ca_file; certifi is promoted
to an explicit dependency.
Closes#1730
Co-authored-by: Isaac
Signed-off-by: manffred-calvosanchez_data <manffred.calvosanchez@databricks.com>
* fix(claude-native): pass a verifying SSL context to wss:// terminal-attach
_websocket_connect opened wss:// terminal-attach connections (the scheme
terminal_attach_url produces from an https workspace base_url) with a bare
default SSL context, so claude-native attach to a remote workspace hit the same
empty-trust-store failure fixed for the tunnels. Route it through
client_ssl_context() for wss:// (ws:// stays ssl=None). Also realign a
ws_tunnel test with the databricks_request_headers rename from main.
Closes#1730
Co-authored-by: Isaac
Signed-off-by: manffred-calvosanchez_data <manffred.calvosanchez@databricks.com>
* chore(deps): record certifi in uv.lock
pyproject.toml promoted certifi to an explicit dependency; add it to the
omnigent package's dependencies and requires-dist in uv.lock so
"uv sync --locked" passes in CI. certifi was already resolved transitively,
so its package entry (with hashes) is unchanged — this only records the
direct dependency edge.
Co-authored-by: Isaac
Signed-off-by: manffred-calvosanchez_data <manffred.calvosanchez@databricks.com>
---------
Signed-off-by: manffred-calvosanchez_data <manffred.calvosanchez@databricks.com>
Co-authored-by: Tomu Hirata <tomu.hirata@gmail.com>
Session-discovered agents start with harness=null (filled lazily on hover
via prefetchAvailableAgentDetails). The fork picker filters candidates with
forkTargetCarriesHistory(a.harness), which returns false for null, so
custom agents were silently excluded from the fork agent dropdown even
though they appear fine in the new-session picker.
Fix: call prefetchAvailableAgentDetails for all agents when ForkSessionForm
mounts (same pattern NewChatDialog uses on dropdown open). The helper is a
no-op for agents whose harness is already known, so re-running on agents
list change is safe.
Adds a test that verifies prefetch is called for a session-discovered agent
(harness=null, sessionId set) on mount.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(web): group archived sessions by date
The archived sessions list in the settings page was a flat
chronological list that became hard to scan. Group sessions under
date headers (Today, Yesterday, Previous 7 days, Previous 30 days,
or month/year for older entries) for easier browsing.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(web): use DST-safe date arithmetic and add grouping tests
Use calendar-based setDate() instead of fixed millisecond offsets for
computing date boundaries in the archived sessions grouping, avoiding
mis-bucketing around DST transitions. Add a Vitest test with a pinned
system clock that verifies all five date group headers render correctly.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(web): share now across grouping, fix test locale/timezone flakiness
- Capture a single `now` in the groupedArchived memo and pass it to
every dateGroupLabel call, avoiding redundant Date construction and
a rare date-rollover inconsistency during iteration.
- Use local-time Date constructors in the test so bucket boundaries
match dateGroupLabel's local-time arithmetic in any timezone.
- Derive the expected month/year label via toLocaleDateString so the
assertion passes under non-English locales.
- Wrap assertions in try/finally so fake timers are always restored.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
---------
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* feat(policy): add detect_loop builtin to catch agent retry loops
The #1 token-waste pattern is an agent retrying the exact same failing
tool call. max_tool_calls_per_session counts total calls but cannot
detect repeated ones. detect_loop tracks recent (tool_name, args_hash)
tuples in session_state and ASKs when the same call repeats N times
within a configurable sliding window, letting the user break the loop.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* address review: use full SHA-256 digest, add e2e tests
- Remove [:16] truncation from _args_hash to use the full 64-char
hex digest, avoiding false-positive collisions from 64-bit space.
- Add YAML → PolicyEngine e2e tests exercising the full roundtrip:
repeated calls trigger ASK, diverse calls pass, window eviction
works, and non-tool_call phases are unaffected.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* address review: guard params, fix docstring, move e2e test
- Clamp window and threshold to minimum 1 so zero/negative values
cannot cause unbounded state growth or always-ASK behavior.
- Add minimum: 1 constraints to both params in the registry schema.
- Fix docstring to describe actual persisted state shape (list of
SHA-256 hex digests, not tuples).
- Move e2e test from tests/runtime/policies/ to tests/e2e/ per
repo convention.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
---------
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* feat(policies): add detect_thrashing builtin context policy
Agents that hit repeated tool errors burn tokens without making
progress. Add a new builtin contextual policy that tracks
tool-result outcomes in a rolling window and fires when the agent
appears stuck — either via consecutive errors or a high error rate
within the window.
Two independent triggers (both configurable, both independently
disableable):
- consecutive_threshold (default 5): fires after N straight errors
- window_error_rate (default 0.8): fires when ≥80% of the last
N results (window, default 10) are errors
Error detection is heuristic (common prefixes like "Error:",
"Traceback", "Permission denied", "fatal:", and JSON {"error": ...}
payloads). No server LLM required, unlike detect_task_switch.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(policies): address review feedback for detect_thrashing
- Fix docstring: "exceeds" → "reaches or exceeds" to match the >= check
- Rename misleading test names (test_below_consecutive_threshold_allows
was actually at-threshold; test_window_rate_allows_below_threshold was
at-threshold)
- Retain max(window, consecutive_threshold) history entries so the
consecutive check still works when window < consecutive_threshold
- Rate check now computes over the last `window` entries (not the full
retained history), and reports window size in the reason message
- Add integration test exercising state accumulation across evaluate
calls through the real policy engine
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(policies): harden detect_thrashing against edge cases
- Validate session_state history as list[int] before use; reset to
empty on corruption instead of raising TypeError.
- Guard against window=0 by using effective_window = max(window, 1)
to prevent division by zero in the rate check.
- Use dataclasses.replace in the integration test to preserve all
original RuntimeCaps fields instead of reconstructing with only
execution_timeout.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(policies): add minimum/maximum constraints to detect_thrashing schema
Add validation bounds to the registry params_schema so invalid config
values fail fast: consecutive_threshold >= 0, window >= 1,
window_error_rate in [0.0, 1.0].
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(policies): use Phase enum in detect_thrashing integration test
Use Phase.TOOL_RESULT instead of the bare string "tool_result" in the
PhaseSelector construction, consistent with other integration tests.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
---------
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(web): clear deleted pinned sessions from the sidebar's Pinned section
The Pinned section reads a sibling ["pinned-conversations"] cache that the
delete mutations' prefix-matched ["conversations"] sweep deliberately skips
(nesting it under that prefix breaks the pin-toggle's cache patch). That
isolation is by design, but it means the delete handlers must drop the row
from the pinned cache explicitly — which they didn't. So deleting a pinned
session removed it from the flat list but left it lingering in the Pinned
section until a full reload.
Mirror the unpin removal in all three delete paths (single-delete onSuccess,
bulk-delete onSuccess, and bulk-delete onError's partial-success branch),
patching the pinned cache in place rather than invalidating for the same
search-reindex-lag reason the list is patched in place.
Co-authored-by: Isaac
* fix(web): keep the sidebar row height stable during delete
The in-flight "Deleting…" status row that replaces an interactive
conversation row used `text-sm py-2` with no height constraint, while
the interactive row uses `sidebar-compact-text h-7 py-0.5`. So starting
a delete didn't just recolor the row — it grew taller and changed font
size, shifting the surrounding list.
Match the deleting row's box metrics to the interactive row (h-7,
sidebar-compact-text font size, otto-sm radius) so the swap only changes
color/opacity.
Co-authored-by: Isaac
* fix(web): keep the sidebar row size stable when editing the title
The inline rename row rendered a `text-sm` (14px) input inside a wrapper
whose `py-1` + `size-7` buttons summed to ~36px, while the interactive
row is `h-7` (28px) with the 13px `sidebar-compact-text` font. So
double-clicking to rename made the row grow taller and bump the font
size, an input visibly larger than the row it replaced.
Match the edit row's box metrics to the interactive row (h-7,
sidebar-compact-text, otto-sm radius) and drop the buttons to icon-xs
(24px) so they sit inside the 28px row, leaving only the muted edit
background to signal the mode.
Co-authored-by: Isaac
* test(e2e): guard pinned-session delete clears the Pinned section
Adds a browser e2e that pins a session (while sitting on `/`, so it isn't
the active chat) and deletes it, asserting the "Pinned" section unmounts
in place — no reload.
Two harness details are load-bearing, and getting them wrong yields a
test that passes even against the buggy build:
- Delete a NON-active pinned session (page on `/`). Deleting the open
session navigates away and refetches; an active session also gets a
WS `removed`-frame reconcile. Either clears the row regardless of the
cache bug.
- Assert the "Pinned" SECTION disappears, not the row's href. While the
delete is in flight the row swaps to a hrefless "Deleting…" status row,
so an href-count assertion flickers to 0 during that transient and
passes spuriously; the section stays mounted until the pinned cache is
actually empty.
Verified it fails (~3s) against a build with the pinned-cache delete
patch removed, and passes with it.
Co-authored-by: Isaac
Collapsed tool runs in the chat transcript now read like the native
CLIs' step summaries ("Ran 1 shell command, read 2 files", "Listed 1
directory") instead of the generic "See N steps". The label is derived
from the folded calls' tool names and arguments in formatToolRunLabel:
- categories: shell / list / read / edit / search, covering omnigent
sys_* tools plus the native harness names (Claude Code Bash/Read/...,
Codex shell/apply_patch, pi & opencode lowercase bash/read/edit/...)
- shell commands that are a bare ls / cat recategorize as directory
listings / file reads, matching the vendor TUIs; codex's login-shell
wrapper (/bin/bash -lc '...') is unwrapped first
- runs of only unrecognized tools fall back to "Called N tools"
- per-step titles added for the native harness tools (Bash prefers the
model-written description, codex shell shows the unwrapped command)
The fold now labels only its own (hidden) contents; the whole-run
count plumbing is gone since the label no longer double-counts the
visible streaming tail.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): show startup spinner when a send relaunches a disconnected runner
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): show a sidebar starting spinner while a session is booting
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): don't rename the wrong session when the sidebar reorders mid-double-click
Double-click rename fired on whichever row received the dblclick event.
Browsers pair the two clicks of a double-click by pointer position and
timing, not element identity, so when the list reordered between the
clicks (an updated_at bump pushing rows around under the cursor) the
second click and dblclick landed on the row that slid into place and
opened rename on it — committing the typed title to a session the user
never aimed at.
Track the last two clicks each row receives and enter rename only when
the row saw both clicks of the pair; a dblclick preceded by a single
recent click means the double-click started on a different row and is
ignored.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): freeze sidebar order under the pointer so single-click actions hit the aimed row
The double-click guard can't help single-event interactions: a right-click
(or kebab click) that lands just after a background updated_at bump opens
the context menu of whichever row slid under the cursor — the menus are
visually identical, so the user renames (or archives, deletes, stops) a
session they never aimed at.
Fix it upstream of any one interaction: while the pointer is inside the
conversation list, pin every row's sort key at its first-seen value so
rows cannot move under the cursor at all. Keys accumulate lazily in
sortByUpdatedAtDesc (covering project folders and pages loaded mid-hover)
and clear when the pointer leaves, snapping the order back to reality.
The active row's frozen key captures its ActiveChatOverride value so
dropping the override mid-hover (clicking another row) can't move it
between the clicks of a double-click either.
Also rebuild the element tree per rerenderSidebar call in the row-actions
test harness — re-rendering the identical element let React bail out
without re-invoking the sidebar, silently ignoring mid-test data swaps.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): hold sidebar order while a rename edit is open, not just while hovered
The order freeze keyed off pointer position alone, but the pointer
naturally drifts out of the sidebar while typing a new title — the hold
released mid-edit and background updated_at churn resumed shuffling rows
around the open input. Moving the edit row's DOM node also blurs the
input, committing a half-typed title.
Rows now report an in-progress inline rename through RowEditHoldContext,
and ConversationList keeps the sort-key freeze active while the pointer
is inside the list OR any rename edit is open. The frozen-key map clears
only once neither hold remains, so the order snaps back on commit/cancel
(or pointer-leave with no edit open).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): engage the rename-edit order hold before paint
A passive effect reports the hold after paint, leaving a one-frame
window — when rename starts with the pointer already outside the list
(context-menu portal) — where a background updated_at reorder could
move and blur the just-mounted input. useLayoutEffect closes the gap.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): make trackpad wheel scrolling work in the terminal view
xterm's built-in wheel-to-mouse-report conversion damps sub-50px pixel
deltas by 0.3x and emits at most one report per DOM event, so macOS
trackpad scrolling over a mouse-tracking TUI (Claude Code, tmux mouse on)
barely moves. Replace it with a custom wheel handler that accumulates
deltas at face value and emits one SGR report per whole line, deferring
to xterm's native handling when the pane program isn't tracking the
mouse (e.g. a plain shell on the control transport).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(terminals): replay pane screen/input modes in the control-mode attach seed
capture-pane records cell contents only, so a TUI that entered the
alternate screen and enabled mouse tracking before the web client
attached (OpenCode, vim — anything that sets modes once at startup)
left the browser xterm believing no tracking was active: wheel events
sent nothing and the terminal view could not scroll until the program
happened to re-toggle its modes. Reconstruct the modes from tmux's pane
flags and replay them around the seed — alt screen before the content
so it never pollutes primary scrollback, mouse tracking/encoding and
DECCKM after the cursor restore.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* test(e2e_ui): pin wheel-to-SGR-report forwarding for mouse-tracking shells
A program in a user shell enables any-motion + SGR mouse tracking and
records its stdin; a slow trackpad-sized wheel gesture over the xterm
must land >=3 wheel-up reports. xterm's damped built-in conversion
yields <=1, so this fails without the accumulating wheel handler
(verified against an unfixed UI build).
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(terminals): harden seed metadata parsing and quote e2e log path
Address review: pad missing/empty tmux mode-flag fields so a flags
anomaly costs only the optional mode replay, never the cursor and
alt-screen state; quote the wheel-log path typed into the e2e shell.
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* test(e2e_ui): type-annotate the wheel test's tmp_path fixture
Signed-off-by: dbczumar <corey.zumar@databricks.com>
---------
Signed-off-by: dbczumar <corey.zumar@databricks.com>
* fix(web): preserve file browser scroll position across session switches
The Files panel's scroll container never tracked its position, so
switching conversations collapsed the list to a loading state and
clamped scrollTop back to 0 with nothing to restore it.
Cache scrollTop per conversation (and per Changed/All view) in a
module-level map — the same pattern FolderTree uses for expanded
paths — restoring it once the view's data is ready, and gating saves
on having restored first so the loading-state clamp can't overwrite
the cached value.
Co-authored-by: Isaac
Signed-off-by: Bryan Qiu <bryan.qiu@databricks.com>
* fix(web): survive the loading clamp when restoring file browser scroll
The first cut restored scrollTop once when isLoading turned false — but
the files queries are disabled (not loading) until the environment query
resolves, so the restore fired against the short placeholder, clamped to
0, and the clamp's scroll event overwrote the cached position.
Gate on data presence instead, re-assert the target via an
animation-frame loop until the container can hold it (or its height
stops changing), and keep saving off until the restore settles.
Also re-sync FolderTree's expanded-paths state from its cache when the
conversation changes without a remount — previously the tree kept the
prior conversation's expanded set, which also skewed content height at
restore time.
Co-authored-by: Isaac
Signed-off-by: Bryan Qiu <bryan.qiu@databricks.com>
* fix(web): keep the open file's scroll position across session switches
The app remembers which file is open per session and re-opens it in the
viewer on switch-back — at the top. The earlier fix only covered the
Files panel list, so what users actually saw (the open file's content)
still reset.
Extract the clamp-surviving restore logic into a shared useScrollRestore
hook (FilesPanel now consumes it) and wire persistence into every viewer
surface, keyed per conversation + path: the Monaco code editor and diff
viewer (via their scroll APIs), the FileViewer content area, the
markdown/notebook previews, and the TipTap markdown editor.
Verified end-to-end in a real browser: Playwright tests scroll, switch
sessions via the sidebar, switch back, and assert the offset returns —
for both the file tree and an open markdown file.
Co-authored-by: Isaac
Signed-off-by: Bryan Qiu <bryan.qiu@databricks.com>
* fix(web): harden scroll restore against async content growth and Monaco clamps
The restore loop gave up as soon as the container's height held still
for one frame — but previews grow in bursts (async syntax highlighting,
image decode, lazy notebook cells), so a single stall stranded the
reader at the top. Replace the giveup with a 1.5s deadline that keeps
re-asserting the saved offset, and settle immediately on wheel/touch/
pointer input so the user is never fought for the scrollbar.
The Monaco surfaces saved onDidScrollChange offsets unconditionally, so
a not-yet-laid-out editor's clamp-to-0 event could permanently overwrite
the cached position. A shared attachEditorScrollRestore helper now
suppresses saves and re-asserts the target until it's reached, the user
scrolls, or the budget expires — the same contract as the DOM hook.
Co-authored-by: Isaac
Signed-off-by: Bryan Qiu <bryan.qiu@databricks.com>
---------
Signed-off-by: Bryan Qiu <bryan.qiu@databricks.com>
Doc sync / Classify and draft docs (push) Has been cancelled
CI / Pytest (databricks) (push) Has been cancelled
CI / Pytest (inner-rest) (push) Has been cancelled
CI / Pytest (integration-mock) (push) Has been cancelled
Lint / Version lockstep check (push) Has been cancelled
Sync OpenAPI to site / Open sync PR on omnigent-site (push) Has been cancelled
UI Preview / notify (push) Has been cancelled
UI Preview / build (push) Has been cancelled
OSS Scorecard / Scorecard analysis (push) Has been cancelled
UI Preview / cleanup (push) Has been cancelled
CI / Pytest (repl-sdk) (push) Has been cancelled
CI / Pytest (runtime-policies) (push) Has been cancelled
CI / Pytest (server-approvals) (push) Has been cancelled
CI / Pytest (server-rest) (push) Has been cancelled
CI / Pytest (slack) (push) Has been cancelled
CI / Pytest (spec-llms) (push) Has been cancelled
web Tests / web test (push) Has been cancelled
CI / Pytest (tools) (push) Has been cancelled
CI / Pytest (misc) (push) Has been cancelled
CI / Pytest (runner-app) (push) Has been cancelled
CI / Pytest (runtime-core) (push) Has been cancelled
CI / Pytest (runtime-harnesses) (push) Has been cancelled
CI / Pytest (server-integration) (push) Has been cancelled
CI / Pytest (stores) (push) Has been cancelled
CI / Pytest (stores-postgres) (push) Has been cancelled
CI / Pytest (stores-mysql) (push) Has been cancelled
CI / Pytest (codex-parity) (push) Has been cancelled
Windows (native) / Windows smoke + unit (push) Has been cancelled
Lint / Pre-commit checks (push) Has been cancelled
CI / Coverage report (push) Has been cancelled
UI Preview / deploy (push) Has been cancelled
Native hook subprocesses (codex, claude, kimi, hermes, cursor) and the pi/opencode
JS extensions have been POSTing directly to the Omnigent server with a baked
30-minute bearer token. After expiry, every hook invocation pays ~1.7s for
credential re-discovery. The relay approach eliminates this class of failure
entirely by removing the server bearer from hook configs.
Changes:
relay handler (claude_native_bridge.py):
Add POST /policies/evaluate to the tool relay HTTP server. The relay
authenticates callers with its existing non-expiring local token and
proxies to the Omnigent server using asyncio.run_coroutine_threadsafe
with the runner's refresh-capable server_client (86400s timeout to
match ASK gate long-polls). session_id is written into tool_relay.json
so hook subprocesses can identify the session without a separate config.
runner/app.py:
Pass server_client and session_id to start_tool_relay so the relay can
serve the /policies/evaluate proxy endpoint.
native_policy_hook.py:
Add read_relay_policy_config(bridge_dir) helper that reads tool_relay.json
and returns (relay_url, relay_token, session_id), and relay_policy_evaluate_url.
Add _RELAY_URL_ENV / _RELAY_TOKEN_ENV constants for env-var harnesses.
hook subprocesses (codex, claude, kimi):
Read tool_relay.json first via read_relay_policy_config; fall back to
direct server call (policy_hook.json / permission_hook.json) when the
relay is not yet up. Remove _PersistingReauth from codex_native_hook.
hermes/cursor hook subprocesses:
Check _OMNIGENT_RELAY_URL / _OMNIGENT_RELAY_TOKEN env vars; fall back to
existing _OMNIGENT_AUTH_HEADERS path when absent.
hermes_native_bridge.py:
Add inject_relay_into_policy_hook which rewrites omnigent-policy-hook.sh
with relay env vars after ensure_comment_relay runs.
orchestration.py:
Wire ensure_comment_relay into _auto_create_pi_terminal (new param) and
inject relay coords into pi config.json and hermes wrapper script after
relay starts. Wire ensure_comment_relay into opencode policy_env via
OMNIGENT_RELAY_FILE. Remove _policy_hook_auth_loop and related refresh
machinery (_register/_unregister_policy_hook_auth, _POLICY_HOOK_AUTH_SESSIONS).
pi extension JS:
Add relayCredentials() that re-reads config.json for relayUrl/relayToken
on each call; evalNativePolicyHttp prefers relay URL and token over direct
server call.
opencode plugin JS:
Add relayCredentials() that re-reads OMNIGENT_RELAY_FILE (tool_relay.json)
on each call; evaluate() prefers relay over direct server call.
pi_native_bridge.py:
Add inject_relay_into_config to write relayUrl/relayToken into config.json.
All harnesses keep a direct-server fallback so sessions started before the
relay is up (first-call race) continue to work. The relay path is taken on
every subsequent call once tool_relay.json is written.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(models): discover Kiro picker catalog
Replace the curated Kiro model picker table with the CLI's JSON model listing so newly released, renamed, or retired Kiro models no longer require an Omnigent source update.
Run discovery on the bound runner, expose it through a dedicated model-options endpoint, and reuse the server's asynchronous single-flight cache so snapshots never block on the CLI process.
Preserve Kiro-provided default, description, context-window, and credit-rate metadata in picker rows, remove four hardcode allowances, and document the discovery boundary.
Tests: 115 Kiro, runner lifecycle, and server snapshot tests; staged pre-commit run; manual validation against kiro-cli 2.10.0 output.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* test(kiro): cover picker discovery failures
Exercise the runner endpoint's retryable 503 path when Kiro CLI model discovery fails so the server keeps its picker cache cold instead of treating failure as an empty successful catalog.
Extend the session snapshot round-trip to verify provider descriptions and rate units survive NativeModelOption's extra-field wire schema alongside context windows and rate multipliers.
Tests: 116 Kiro, runner lifecycle, and snapshot tests passed. Live kiro-cli 2.15.1 discovery returned nine models with auto as the sole default. Targeted pre-commit passed.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(kiro): narrow discovered picker contract
Remove the unused kiro_base_model_options compatibility alias because its old pure lookup contract became a blocking CLI subprocess and no production caller remains.
Stop emitting isCurrent for Kiro because the CLI discovery response does not provide current-session state and the Web picker derives the selected row from model_override.
Cover missing and mismatched CLI defaults in the discovery mapper and verify that the Web picker falls back to its Default sentinel, leaving Kiro responsible for choosing the actual default. Refresh the Kiro picker E2E fixture and wording to match live discovery.
Tests: 118 focused Kiro/runner/snapshot tests; 4,705 Web tests; targeted pre-commit checks.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
First half of the runner launch seam. Wires the provider's auto_create_terminal
field (declared since 1.1, never dispatched) and collapses the 8 uniform
create-session launch arms in runner/app.py onto it. Behavior-preserving.
- orchestration: add NativeLaunchContext (flat dataclass of the inputs the 11
builders may need, incl. claude's closures), PreLaunchResult (skip /
force_recreate / needs_terminal for the special arms in 1.5b-ii), 11 thin
_launch_<x>(ctx) adapters that unpack the context and call the unchanged
_auto_create_<x>_terminal builder with that harness's exact kwarg subset, and
the shared shell _launch_native_terminal(harness, ctx, *, ensure_locks,
pre_launch=None, resolve_agent_spec=None). The shell runs the lock /
existence-check / pending+error-event mechanics every arm shared and resolves
the adapter via resolve_hook(provider, "auto_create_terminal").
- Option A (adapters, builders unchanged) keeps the 21 direct-call builder tests
intact. agent_spec is resolved lazily via resolve_agent_spec inside the create
block, preserving each arm's error semantics (pi unwrapped; cursor/opencode/
kimi swallow OmnigentError via _resolve_session_agent_spec_or_none; the rest
pass no resolver).
- harness_plugins: repoint auto_create_terminal to omnigent.runner.native:_launch_<key>.
- app.py: the 8 uniform arms (pi, cursor, kiro, opencode, goose, hermes, qwen,
kimi) become one _launch_native_terminal call each, picking the per-harness
lock dict (kept app-scope so session cleanup can pop by name). Net -256 lines.
- qwen's launch-error label is now "Qwen Code" (uniform display_name) vs the
former lowercase "qwen" — cosmetic; no test asserted the literal.
Deferred to 1.5b-ii: the 3 special arms (claude/codex/antigravity) and the
turn-path opencode cold-boot, which still use the direct builders.
Tests: unit-cover each adapter's kwarg subset and the shell's branches
(create / existing-skip / force-recreate teardown / skip+needs_terminal /
start-error event / lazy-spec-only-on-create / non-native None). The workflow-
init HTTP suite exercises the real launch path for the uniform arms and stays
green. Pre-existing codex gateway-env failures in events_lifecycle are unchanged
(verified identical on clean main; codex arm untouched here).
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
Remove the release-specific smart-routing model table and require the runner worker catalog for routing candidates. When discovery is unavailable, leave the harness on its provider-resolved default instead of selecting a stale fallback.
Order catalog candidates by normalized provider-relative cost tiers while preserving catalog order as the tie-breaker, and express the built-in judge rubric through stable fast, balanced, and powerful intents rather than vendor model-name tiers.
Apply the same discovery-only rule to sys_advise_models, ratchet eight hardcode allowances, and document the remaining wire-compatibility exclusions as a separate migration boundary.
Tests: 67 focused routing/session tests; staged pre-commit run.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* feat(models): resolve runtime defaults from catalogs
Adapt MLflow provider listings into normalized resolver candidates with tri-state capability metadata, context windows, provider-relative cost tiers, and deterministic family filtering.
Replace release-specific defaults across workflow ucode routing, SDK executors, Databricks execution, and Claude/Codex/Pi/OpenCode native launch paths. Explicit request, spec, ucode, and provider-configured models continue to win; unresolved defaults now use the active provider catalog and fail clearly when discovery has no compatible model.
Improve model-version sorting so provider prefixes, dates, endpoint sizes, and unrelated numeric families do not distort catalog order. Ratchet ten obsolete hardcode allowances and document the runtime migration boundary.
Tests: 168 catalog/workflow tests; 519 executor tests; 449 native tests; staged pre-commit run.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): honor overrides before defaults
Apply per-session and CLI model overrides to the effective executor spec before spawn-environment builders attempt provider default resolution. This keeps explicit request values authoritative when catalog lookup is unavailable.
Preserve an explicit OMNIGENT_MODEL value when --harness selects the runtime, and allow model-only E2E overrides when the YAML owns harness selection. Add deterministic fixture models to unrelated tests so catalog-disabled CI does not depend on discovery.
Tests: 8 catalog-disabled CI regressions; 284 broader CLI/runtime/runner tests; staged pre-commit including the hardcoded-model lint.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): preserve catalog default policy
Route default-intent catalog resolution through the existing general-purpose selection policy after family filtering. This retains specialty-model exclusion and provider pins while leaving non-default intents on metadata ranking.
Require dynamically discovered Databricks defaults to use gateway-routable databricks-prefixed ids, report actionable catalog misses to direct executor callers, and model context capacity from max input tokens rather than input plus output budgets.
Add regression coverage for constrained defaults, lagging provider pins, OpenAI specialty variants, Databricks Claude/OpenAI routing, and context-window normalization.
Tests: 85 focused catalog/provider tests passed; 150 broader tests produced 149 passes plus the documented ambient Claude-login failure. Live Databricks catalog verification found 14 Claude and 16 OpenAI entries, all gateway-prefixed. Pre-commit passed all relevant hooks; repository-wide web-prettier and stale routing protobuf checks remain baseline failures.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): offload catalog discovery
Run cold catalog resolution on the existing dedicated thread helper from async Codex, Databricks, Open Responses, OpenAI Agents, and Pi turn paths. Model-less first turns can now wait for remote discovery without blocking the shared event loop for the catalog timeout.
Keep explicit and configured model precedence synchronous and unchanged. Make Pi's internal model resolver async so its Databricks fallback follows the same non-blocking boundary.
Add a regression that verifies catalog discovery executes outside the event-loop thread and update Pi resolver tests for the async contract.
Tests: 405 affected executor tests passed. Targeted pre-commit passed, including formatting, Ruff, and the hardcoded-model lint.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(models): offload Claude catalog lookup
Move the remaining Claude SDK Databricks catalog fallback onto the dedicated thread helper so a cold remote lookup cannot block the async turn loop.
Restore direct Pi coverage tying a catalog-selected Databricks default to dynamic models.json registration. This preserves the prior unknown-model invariant even when the selected gateway id is newer than Pi's curated static entries.
Tests: 251 Claude SDK and Pi executor tests passed with the documented macOS path-canonicalization test deselected. Targeted pre-commit passed, including Ruff and the hardcoded-model lint.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(harness): route native spawn-env through the provider seam (PR 1.5a)
Collapse the two near-identical 11-arm native spawn-env dispatch chains in
runner/app.py (create-session ~2567 and dispatch ~6092) onto the provider seam.
Each block becomes one guarded call to a registry-driven helper; net -171 lines
in app.py. Behavior-preserving — every native harness produces the identical
spawn env before/after.
- harness_plugins: populate `spawn_env_builder` on all 11 built-in providers
(uniform `omnigent.<key>_native_bridge:build_<key>_native_spawn_env`) and add
a `bridge_id_label_key` field, set to `omnigent.<key>_native.bridge_id` for
the three label-based harnesses (codex/opencode/antigravity). The label key is
derived (not imported) to keep harness_plugins import-light; a test pins the
derivation against the real bridge constants.
- runner/native/orchestration: add `_resolve_native_spawn_env(harness, session_id,
*, server_client, optional_labels)`. It resolves `provider.spawn_env_builder`
and handles the three shapes — bare (session id only), label (bridge id from
`bridge_id_label_key`), and two named specials: claude (bridge id via the
runner helper with a server-side fallback) and hermes (writes its policy-hook
config before building). Returns None for non-native harnesses so the caller
keeps its SDK spawn env. Re-exported via runner/native/__init__.
- runner/app: both blocks now call the helper; the per-harness bridge imports and
label-key reads are gone.
The two special-cases (claude/hermes) stay named branches in the helper rather
than fully data-driven provider fields — their only consumers are single call
sites, and 1.5b's NativeLaunchContext will reshape the right calling convention.
Tests: extend the provider-paths-resolve + required-hooks tests to cover
spawn_env_builder; pin bridge_id_label_key against the real constants; add
`_resolve_native_spawn_env` unit coverage for all four shapes + the non-native
None path. The existing workflow_init codex-bundle-dir spawn-env test (the
end-to-end behavior-preservation proof) stays green unchanged.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* test(harness): hoist spawn-env test imports to module level
Move the per-test `_resolve_native_spawn_env` and
`CODEX_NATIVE_BRIDGE_ID_LABEL_KEY` imports (added in 1.5a) up to the module
import block. No behavior change; test-only cleanup.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
## Related issue
N/A
## Summary
- `@testing-library/jest-dom` never declares `vitest` as a (peer) dependency.
Under pnpm's store layout, jest-dom's `declare module "vitest"` matcher-type
augmentation can't resolve `vitest`, so it silently fails to merge and `tsc`
loses every DOM matcher (`toBeInTheDocument`, `toHaveClass`, …) — even though
they register fine at runtime. See vitest-dev/vitest#10411.
- Declare the missing peer via pnpm `packageExtensions` so pnpm links `vitest`
into jest-dom's scope and the augmentation resolves. This is a root-cause fix
at the dependency layer — no hand-written type shim needed.
- Note: `type-check` still has unrelated pre-existing errors and is not yet
gated in CI; this fix only removes the jest-dom matcher category.
## Test Plan
- `pnpm install --frozen-lockfile --filter web` — lockfile stays consistent.
- `pnpm --filter web run type-check` — jest-dom matcher errors drop from 1589
to 0 (remaining errors are unrelated and pre-existing).
- `pnpm --filter web run test` (e.g. `src/shell/WorkspacePanel.test.tsx`) —
15/15 pass under Node 22; runtime is unaffected.
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [x] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Verified by comparing `pnpm --filter web run type-check` jest-dom error counts
(1589 → 0) and running the existing vitest suite (unaffected). The change is
dependency-resolution config only, with no new runtime code to test.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
The module docstring only showed the session policy REST API, implying
CEL policies can't be declared statically. Both static paths work and
are now shown: config.yaml policies (handler + factory_params, parsed
by omnigent.inner.loader) and bundled agent specs (guardrails.policies
with a function {path, arguments} mapping, parsed by
omnigent.spec.parser — which does not read factory_params). Verified
both forms against their parsers.
Signed-off-by: Harry Su <tiecheng.su@robinhood.com>
* fix: add codex_cli_version to fake app-servers in tests; fix ruff format
- Add codex_cli_version = None to all _FakeCodexAppServer classes so they
satisfy the new attribute read in the orchestration bypass_hook_trust gate
- Collapse the multiline boolean in orchestration to satisfy ruff format
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: symlink hooks.json into private CODEX_HOME so user hooks fire
hooks.json was never symlinked, so user hooks declared there were silently
ignored in private sessions. Add it to _CODEX_HOME_GLOBAL_INSTRUCTION_FILES
so it's symlinked in full sessions but skipped in minimal_config (title
worker) mode. Trust is no longer a concern since --dangerously-bypass-hook-trust
is passed to runner-owned TUI sessions.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: merge user hooks.json into policy hooks file instead of clobbering symlink
_write_codex_policy_hooks_file was using os.replace() which destroyed the
hooks.json symlink created by _populate_codex_home_config, silently dropping
all user hooks. Now when the path is a symlink, we read the user's hooks,
merge them after the policy hooks for each event (plus any user-only events),
remove the symlink, and write the merged payload as a regular file.
User hooks from ~/.codex/hooks.json now fire in private sessions.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* style: collapse _merge_user_hooks signature to one line (ruff format)
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
Scopes the listing to system.ai models only via the parent filter and
raises the result cap to 1000, matching the recommended API call at
/ajax-api/2.1/unity-catalog/model-services?max_results=1000&parent=schemas%2Fsystem.ai.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(web): scale conversation sidebar text with the font-size setting
The sidebar's compact text was pinned to a fixed `--sidebar-font-size:
13px`, so the Appearance font-size setting only moved the surrounding
rem-based padding while the text stayed at 13px. Express the variable in
`rem` (0.8125rem = 13px at the 16px default) so it rides the root
font-size, which already folds in `--ui-font-scale` and the mobile bump.
Drop the explicit `line-height` on `.sidebar-compact-text`: single-line
rows use fixed height + flex centering (line-height inert), and the two
line-clamped previews now inherit the root's unitless 1.5, which scales
with the text for free.
Co-authored-by: Isaac
* test(e2e-ui): regenerate visual baselines
---------
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
* feat(models): add intent resolver contracts
Define stable model intents and provider-neutral metadata for capabilities, context windows, cost tiers, and wire APIs. Capability support is tri-state so incomplete provider listings cannot be mistaken for positive support.
Add deterministic resolution precedence for explicit choices, configured defaults, live catalogs, and documented static fallbacks. Catalog order remains the tie-breaker, while provider-specific preference policies can override ranking without changing callers.
Expose normalized metadata through model catalog entries and payloads without changing any executor or routing defaults in this slice.
Tests: 108 focused resolver, catalog, and smart-routing tests; staged pre-commit hooks.
Part of #3426
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(models): keep resolver intents caller-backed
Limit the public model intent vocabulary to default, fast, balanced, and powerful because those are the only purposes represented by current callers.
Express tool use, image generation, structured output, and similar requirements through explicit capabilities instead of speculative intent-to-capability mappings. Remove the unused large-context ranking path and update resolver tests and migration guidance accordingly.
Tests: uv run --no-sync pytest -q tests/test_model_resolver.py tests/test_model_catalog.py tests/server/test_smart_routing.py; pre-commit run
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* refactor(models): complete wire API contract
Cover every model-endpoint request shape implemented by the provider adapters by adding Bedrock Converse and naming Gemini generateContent explicitly. Keep native CLI and ACP transports outside the model wire protocol vocabulary.
Clarify that explicit model overrides bypass compatibility constraints, intent tiers are best-effort ranking preferences, and uncatalogued explicit resolutions have unknown family and metadata. Add regression coverage for those semantics and for the complete wire API vocabulary.
Tests: uv run --no-sync pytest -q tests/test_model_resolver.py tests/test_model_catalog.py tests/server/test_smart_routing.py tests/llms/test_openai_adapter.py tests/llms/test_anthropic_adapter.py tests/llms/test_gemini_adapter.py tests/llms/test_vertex_adapter.py tests/llms/test_bedrock_adapter.py tests/llms/test_databricks_adapter.py; pre-commit run
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* docs(harness): revise Phase 1 estimates from runner exploration
Reading the runner dispatch surface (not guessing) changed the shape of the
remaining work, so update the proposal's estimates and plan:
- Split PR 1.5 into a serial runner sub-stack: 1.5a spawn-env (bounded, the
first measurement), 1.5b launch (the epicenter — _auto_create_<x>_terminal
has 11 divergent signatures, so the seam passes a NativeLaunchContext to a
uniform provider.auto_create_terminal(ctx) adapter with pre_launch hooks,
not a single positional call), 1.5c terminal-route.
- Re-scope 1.6 interrupt/stop upward (Med -> Med-High, 2d -> 3-4d): every
handler closes over app-scope state (server_client, resource_registry,
_publish_event, module dicts), so extraction needs a DI context, not a move.
- Revise totals: Phase 1 ~17-25 -> ~20-29 eng-days; overall ~26-37 -> ~29-41
across ~12 -> ~14 PRs; critical path rewritten to the serial runner chain.
- Add a Calibration subsection recording the learning from 1.1-1.3 (additive
PRs come in under estimate; the real cost is test-shape churn; the runner is
the back-loaded risk) and settle the "signature uniformity" open question
with the confirmed finding.
Docs-only; no code paths affected.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* docs(harness): record harness-bench compatibility with native plugins
The harness bench's selection + driver layer is already registry-driven:
manifest.py auto-adds every NATIVE_TUI capability as a BenchProfile and the
NativeTuiDriver is selected generically, so a community native plugin
enumerates and gets a profile with zero bench edits. Record the two remaining
gaps and where they close:
- Provisioning needs registry-driven agent seeding — closed for free by PR 1.7
(the native driver provisions against a pre-seeded <harness>-ui agent).
- Tool-call probe metadata is hardcoded (_NATIVE_TOOL_PROVOCATION) — fold
optional shell_tool_name / shell_tool_prompt capability fields into PR 1.8 so
the probe reads off the registry; until then those probes skip (non-fatal).
Add a "Harness bench compatibility" subsection, extend 1.8's scope with the
tool-probe fields, and give 2.4 a benchable acceptance criterion (the example
plugin runs `python -m tests.harness_bench --harness <plugin> --live` green).
No new phase or standalone bench-migration PR.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* refactor(harness): route native resume through the provider seam (PR 1.3)
Collapse the two hand-written native-resume dispatch chains onto
native_dispatch.resolve_hook_for_key(key, "run_native"):
- resume_dispatch._dispatch_wrapper: 10 `if native_agent.key == "<x>"` arms →
one resolved call.
- chat._redirect_native_resume_if_needed: 6 arms + the 6
_run_<x>_native_resume_redirect helpers → one resolved call that derives the
redirect notice from the agent row (wrapper_name == agent.harness,
native_command == agent.key, both verified equal to the old literals) and
passes auto_open_conversation. Deletes the helpers.
Behavior change (intended fix): routing through the seam covers all 11 natives,
closing two latent coverage gaps that double-posted each user turn (the exact
hazard the cursor/kimi docstrings warned about):
- chat redirect covered only 6 of 11 — goose/hermes/antigravity/qwen/opencode
resumes fell through to the Omnigent REPL.
- resume_dispatch covered only 10 of 11 — opencode fell through the same way.
No test pinned either old fall-through; added a chat goose regression test, a
chat unknown-wrapper → False test, and a resume_dispatch opencode test.
Also:
- native_dispatch.resolve is no longer cached — dispatch happens once per
resume/launch/seed, import_module already caches the module, and caching the
resolved attribute silently defeats monkeypatch.setattr("...:run_x", ...),
which the resume/CLI tests rely on. Dropped reset_resolve_cache_for_tests.
- Normalize the cli.py _NativeTerminalDispatchSpec launch table to
args_param="extra_args" (finishing 1.2's spelling migration into the launch
hub) and update the tests that captured the old <x>_args kwarg.
Net -261 lines. Full resume/chat/cli/native suites green; new-failure delta vs.
the clean tree is zero.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* docs(harness): record PR 1.3 in the progress ledger
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
- Gate both approval event and resolve URL paths at owner access
- Prevent shared editors from authorizing tools using owner credentials
Refs #2150
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* refactor(harness): normalize native launcher pass-through args (PR 1.2)
The 11 run_<x>_native launchers each spelled their pass-through arg
differently (claude_args, pi_args, ...). The provider seam needs one uniform
spelling to call them generically. Introduce extra_args as that spelling and
keep <x>_args as a back-compat alias.
- Add native_terminal.normalize_extra_args(): reconciles extra_args vs the
legacy <x>_args alias — extra_args wins, the legacy alias emits a
DeprecationWarning (removal targeted for 0.9.0), neither yields ().
- Give all 11 run_<x>_native entry points a keyword-only extra_args and make
<x>_args an optional deprecated alias, normalizing at the top of each body
so the deep internals keep using the existing local variable unchanged.
- Migrate the internal callers (resume_dispatch ×10, chat resume-redirect ×6,
cli_native ×11) to extra_args so nothing in core trips the new warning; the
alias exists purely for external back-compat.
- Tests: unit-cover the four normalize_extra_args branches. Existing native
tests that still call <x>_args= now double as back-compat coverage.
No behavior change: with default warning filters the full native + hub suite
is green (verified the failure set is byte-identical to the clean tree; the
handful of red tests are pre-existing gateway-env artifacts unrelated to this
change).
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* docs(harness): record PR 1.2 in the progress ledger
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
The scheduled server-compat matrix builds its default version set from
all tags, filtering only rcN. Dev/pre tags are snapshots of main, so
main-vs-them cells add no compat signal, and under the 256-job matrix
cap they evict the oldest final releases — the coverage the workflow
exists for. A stray v0.4.0.dev0 tag is already in the live matrix
today, and a nightly prerelease lane would add ~25 such tags a month.
Explicit VERSIONS dispatch overrides still accept prerelease tags.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(repl): don't report an Omnigent credential for ACP-backed sessions
acp / acp:<slug> / goose / qwen aren't in _HARNESS_FAMILY, so
default_provider_for_harness treats them as unmapped and falls through to
the configured anthropic/openai default. describe_active_credential then
hands back that provider's default_model and credential source, and both
the /model readout and the startup header render it as the active model.
But an ACP agent carries its own auth and picks its own model — the
executor only forwards a model at session/new when send_model_in_session_new
is set. So `omnigent run --harness acp:<agent>` confidently names a model
and an API key the session never touches.
Declines these harnesses at the resolver rather than the readout, so the
startup header stops fabricating too. The predicate reads the declared
capability record (ACP_SUBPROCESS + OWN_AUTH) instead of a hardcoded list,
so community ACP plugins are covered without further edits.
Signed-off-by: apeltekci <andrew@peltekci.com>
* fix(repl): scope the own-auth credential decline to acp/goose and keep overrides visible
The own-auth predicate wrongly included qwen: a harness mapped in
_HARNESS_FAMILY is provider-routed at spawn (_build_qwen_spawn_env injects
the configured openai-family default via
configure_agent_harness_with_provider, and QwenExecutor exports
OPENAI_BASE_URL / OPENAI_API_KEY / OPENAI_MODEL into the qwen subprocess —
see test_qwen_uses_openai_global_default), so its readout naming that
provider was truthful, and declining it fabricated "own auth" in the other
direction. The decline now applies only to unmapped ACP_SUBPROCESS +
OWN_AUTH harnesses (acp/acp:<slug>, goose, unmapped community ACP plugins).
The predicate is public now, so the REPL stops importing a private name,
and the manual acp:<slug> split is gone (canonicalize_harness already folds
it).
The own-auth readout also no longer claims an Omnigent-side /model override
does not reach the agent — model_env_keys() covers acp and goose, the
process manager respawns on a model change, and goose applies the override
as GOOSE_MODEL — and a live override is shown instead of hidden.
Tests: the resolver-level case now uses a key-kind openai default, the kind
the unmapped fallback actually fabricated (a subscription default was
already declined before the fix, so the previous case pinned nothing), and
new cases pin override visibility and qwen's provider-routed readout.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: apeltekci <andrew@peltekci.com>
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Co-authored-by: Dhruv Gupta <dhruv.gupta@databricks.com>
DELETE /auth/users/{user_id} checked whether another admin existed and
deleted the target in two separate, unlocked transactions. Two
concurrent deletes of two different admins could each observe the
other as the remaining admin, both pass, and both apply, leaving
the deploy with zero admins and no in-app recovery path.
Lock the current admin set before counting it (BEGIN IMMEDIATE on
SQLite, SELECT ... FOR UPDATE on other dialects) so the check and
the delete happen in one transaction. A concurrent delete of a
different admin now blocks until the first commits and re-observes
the up-to-date count instead of a stale one.
Signed-off-by: Nick Chakre <nickchakre18@gmail.com>
## Related issue
N/A
## Summary
Bump the Android module's `compileSdk` and `targetSdk` from 35 to 36 to meet
Google Play's requirement that apps target API level 36 by August 30, 2026.
This required updating the full Android toolchain:
- AGP 8.6.1 → 9.1.1 (AGP 9 has built-in Kotlin support)
- Gradle wrapper 8.9 → 9.3.1
- Gradle Play Publisher 3.12.1 → 4.0.0
- AndroidX dependencies to versions compatible with compileSdk 36 (e.g.,
`androidx.core` 1.18.0, `androidx.activity` 1.12.4, `androidx.webkit` 1.15.0)
- Robolectric 4.14.1 → 4.16.1
The `org.jetbrains.kotlin.android` plugin is no longer applied because AGP 9
bundles Kotlin compilation support. Build-script helper tasks that previously
used the Gradle `exec { }` DSL were switched to `ProcessBuilder` to stay
compatible with the new Kotlin/Gradle DSL scope, and `android.sdkDirectory`
was replaced with `androidComponents.sdkComponents.sdkDirectory`.
## Test Plan
Ran the full local Android build pipeline:
```bash
cd web/android
./gradlew :app:assembleDebug :app:lintDebug
./gradlew :app:bundleRelease
./gradlew :app:assembleDebugAndroidTest
```
All completed successfully and produced a debug APK, release AAB, and androidTest
APK with zero lint errors.
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [x] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Verified by running `:app:assembleDebug`, `:app:lintDebug`, `:app:bundleRelease`,
and `:app:assembleDebugAndroidTest` locally. The existing CI `android-bundle.yml`
workflow uses the Gradle wrapper and JDK 17, both compatible with the updated
toolchain.
## Changelog
Android app now targets Android 16 (API 36) to stay compliant with Google Play's
latest target API level policy.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
The repo migrated to a pnpm workspace (pnpm-workspace.yaml and
pnpm-lock.yaml at the root, packageManager: pnpm@11.15.1) but
setup.py's _build_web_ui still shelled out to 'npm install' / 'npm
run build' from inside web/. That path looked for a package-lock.json
that doesn't exist there (the lockfile is pnpm-lock.yaml at the
workspace root), so npm re-resolved from package.json alone and
hard-failed on the @lobehub/fluent-emoji@4.1.0 peer range
(react@^19 vs the pinned react@18.2.0) with ERESOLVE.
Migrate _build_web_ui to pnpm, matching deploy/databricks/build.sh
and the CI workflows (.github/workflows/e2e-ui.yml):
- Resolve pnpm via shutil.which('pnpm'), falling back to
'corepack pnpm' (corepack ships with Node 22+ and auto-pins the
version from package.json's packageManager field).
- Run from the workspace root (cwd=root), not web/, so pnpm uses
the committed pnpm-lock.yaml.
- 'pnpm install --frozen-lockfile --filter web' then
'pnpm --filter web run build' — exactly the CI commands.
--frozen-lockfile guarantees the build is reproducible and
resolves @lobehub/fluent-emoji against react@18.3.1 under the
workspace's strictPeerDependencies: false, avoiding the peer
conflict that broke npm.
Also enforce the Node.js 22 LTS floor up front via a new
_require_node_22 helper that fails fast with a dedicated, actionable
message if 'node' is missing or reports < 22 — instead of failing
deep inside the toolchain with an opaque error.
All existing skip/force env vars are preserved:
OMNIGENT_SKIP_WEB_UI=true (opt out), OMNIGENT_BUILD_WEB_UI=1
(force rebuild), skip-when-bundle-exists, skip-when-web-absent.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
## Related issue
N/A
## Summary
- Split `SandboxLauncher` into a layered hierarchy: `SandboxLifecycle`
(lifecycle + capabilities), `SandboxExecTransport` (run/put/stream/exec),
`SandboxHostLauncher` (abstract start_host), and `ExecModelHostLauncher`
(default start_host + run_background + materialize_workspace).
- `SandboxLauncher` is now a backward-compat alias for `ExecModelHostLauncher`.
- Migrated Kubernetes to inherit `SandboxHostLauncher` directly — it no
longer needs a fake `run()` that raises; the entrypoint-as-host model
(Pod boots running the host) has no exec transport at all.
- All 8 providers now declare an explicit `capabilities` property instead
of relying on class-var derivation.
- Updated the registry's `isinstance` guard to check `SandboxLifecycle`
(the common base) so both exec-model and entrypoint-as-host providers pass.
- Updated the Kubernetes test that asserted `run()` raises to assert the
method does not exist instead.
## Test Plan
```bash
uv run pytest tests/onboarding/sandboxes tests/sandbox tests/cli/test_cli.py tests/server -k managed -q
pre-commit run --files <all changed files>
```
All 780 selected tests pass and pre-commit is clean.
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [ ] Manual verification completed
- [x] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Existing provider and CLI tests pass unchanged, confirming backward
compatibility. The Kubernetes test was updated to reflect that `run()` no
longer exists on the launcher. The registry test was updated for the
`SandboxLifecycle` guard message.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
A conversation id pasted with surrounding punctuation (e.g. a trailing
period) crashed `omni resume` with a raw StatementError traceback from
the local store's Uuid16 bind. Strip the punctuation a paste drags
along — none of it can be part of a valid id — and resume the id the
argument contains, canonicalized to bare hex so downstream consumers
never see a legacy spelling. Error only when no valid id remains.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Add a `./gradlew recordScreenshots` task that captures four real-WebView
screenshots of the Android shell on a device/emulator, with zero manual
setup — Gradle starts and stops both the Vite dev server and an isolated
omnigent backend automatically.
Screens captured (app/build/screenshots/):
- server_select.png — native ConnectActivity (server-entry screen)
- home.png — SPA landing page (sidebar closed)
- session_list.png — SPA home with sidebar drawer open (?sidebar=open)
- session.png — session/chat page with a real seeded user message
How it works:
- startBackendServer: launches `omnigent server` in a throwaway mktemp
data dir (OMNIGENT_DATA_DIR/CONFIG_HOME/DATABASE_URI isolated from
~/.omnigent, no-auth on loopback), pre-registers examples/kimi_hello.yaml.
- seedDemoSession: POST /v1/sessions with an initial user message so the
session screenshot has real content.
- startWebDevServer: launches `node vite --host 127.0.0.1 --port 5173`
directly (avoids spawning npm/pnpm whose grandchild is hard to kill),
reuses an existing server if present. Vite proxies /v1 to the backend.
- Per screen: pm clear + pre-grant POST_NOTIFICATIONS, then drive the real
ConnectActivity → MainActivity flow via UI Automator (am instrument, not
AGP's connectedDebugAndroidTest which auto-uninstalls and deletes the
screenshot before we can pull), then adb pull the PNG.
- stopWebDevServer / stopBackendServer: tear down both + clean temp dir.
The test (ScreenshotTest.kt) is pure UI Automator (out-of-process, black-box):
it launches the app from the launcher, types the server URL (base + route
path) into ConnectActivity, taps Connect, waits for the floating switch pill
as the "shell is up" signal, then captures via UiDevice.takeScreenshot. The
session-list screen uses the ?sidebar=open query param (AppShell reads it on
mount to open the conversation drawer) since uiautomator can't see inside the
WebView to tap the toggle button.
Dependencies added (pinned to the AGP 8.6 / compileSdk 35 toolchain):
androidx.test:runner 1.6.2, :rules 1.6.1, ext:junit 1.2.1
androidx.test.espresso:espresso-core 3.6.1
androidx.test.uiautomator:uiautomator 2.4.0
Also sets testInstrumentationRunner = AndroidJUnitRunner.
Usage:
ANDROID_SERIAL=emulator-5554 ./gradlew recordScreenshots
open app/build/screenshots/*.png
Requires an emulator or unlocked device. The backend/Vite are fully managed
— no separate terminals needed.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
* fix(release): harden finalize, cut, and homebrew against the 0.7.0-cycle failure modes
- finalize: docs sweep is advisory (never blocks publish), untagged drafts
are rebound automatically, tag input is normalized
- release: bump-main gates in shell so CLI-dispatched boolean inputs cannot
silently skip the post-release main bump
- update-homebrew: defer inside PyPI's 24h --uploaded-prior-to window and
add a nightly catch-up that no-ops when the formula is current
- uv.lock: gitpython 3.1.50 -> 3.1.55 (clears 8 OSV advisories that tripped
the Security Scan on every lock-touching PR)
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(images): serialize image builds and raise the build timeout to 120m
At the v0.7.0 cut the rc1 (21:51) and final (21:57) tag builds ran
concurrently under SHA-keyed concurrency, raced each other's layer cache
cold, and the final build died on the 60m job timeout — no v0.7.0 or
latest images until a manual re-run a day later. A single serialized
group lets the later build reuse the earlier one's layers; 120m gives a
genuinely cold build headroom.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
N/A
This is the final npm -> pnpm migration step for the OSS repo.
- Adds `editors/vscode` and `deploy/cloudflare` to `pnpm-workspace.yaml` so
they use the root `packageManager: pnpm@11.15.1` and the shared
`pnpm-lock.yaml`.
- Removes the per-package `package-lock.json` files and deletes the now-obsolete
`scripts/normalize_package_lock_registry.py` hook/script.
- Merges the three remaining categories of build-script approvals into
`pnpm-workspace.yaml` (`@vscode/vsce-sign`, `esbuild`, `keytar`, `sharp`,
`workerd`) so `pnpm install` works at the workspace root.
- Migrates VS Code and release workflows to `setup-pnpm`:
- `.github/workflows/vscode-extension-release.yml`
- `.github/workflows/vscode-release-pr.yml`
- `.github/workflows/release-omnigent.yml`
- Updates the lockfile regen workflows to refresh `pnpm-lock.yaml` instead of
the old web-only `package-lock.json`:
- `.github/workflows/oss-regenerate-and-smoke.yml`
- `.github/workflows/oss-regen-on-comment.yml`
- Updates `editors/vscode/README.md`, `editors/vscode/PUBLISHING.md`, and
`deploy/cloudflare/README.md` to reference pnpm commands.
- Removes the deprecated `.github/actions/setup-node` composite action.
- `pnpm install --frozen-lockfile --filter omnigent-vscode` passes locally.
- `pnpm install --frozen-lockfile --filter omnigent-cloudflare` passes locally.
- `uv run pre-commit run --all-files` passes (after dropping the package-lock
registry hook).
- Inspected remaining `npm install` occurrences in workflows; the only survivors
are transient agent CLI installs (`@anthropic-ai/claude-code`,
`@openai/codex`) that are intentionally not tracked in the lockfile.
N/A
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
Verified the new workspace packages install from the frozen pnpm lockfile and
that the pnpm-only lockfile regen scripts produce a valid lock. The VS Code
workflow commands were checked against the package names/filters from
`pnpm-workspace.yaml`.
N/A
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
The linux_bwrap sandbox mounts a fresh procfs under --unshare-pid, but a
Lakebox microVM masks /proc so that mount returns EPERM and the sandbox
fails to start. That blocked linux_bwrap — and the L7 egress management
built on top of it — on the Lakebox backend.
Bind the existing /proc instead of mounting a fresh one, but only on
outer sandbox backends known to be safe for it (allow-list: lakebox).
The backend is read from OMNIGENT_HOST_SANDBOX_BACKEND when set, else
autodetected via the /run/lakebox marker. Everywhere else the fresh-proc
mount and its fail-closed behavior stay unchanged.
Binding /proc exposes the outer process list and world-readable per-proc
files (cmdline/comm/stat/status). The retained user namespace still
blocks ptrace-gated files (environ/mem/maps/fd) and --unshare-pid still
contains signalling, so the leak is acceptable on a single-tenant
Lakebox microVM.
Signed-off-by: Thomas Garnier <6202935+mxatone@users.noreply.github.com>
The kimi forwarder mirrored wire content but never posted an
external_session_status edge — the only native forwarder that didn't
(claude/codex/opencode/cursor all do). A kimi sub-agent therefore finished,
delivered its answer to the transcript, and left the parent waiting on it
forever: _mark_subagent_terminal_and_wake was never reached, so no result
ever landed in the parent's inbox.
kimi's wire has no turn.end row; its agent loop steps while step.end carries
finishReason 'tool_use' and stops on 'end_turn' (1:1 with turn.prompt across
every recorded session). Map that edge to external_session_status: idle,
carrying the turn's final assistant text — the runner delivers an empty
result when an idle edge forwards none.
Signed-off-by: apeltekci <andrew@peltekci.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* feat(harness): add NativeHarnessProvider seam foundation (PR 1.1)
First, additive step of Phase 1 of the modular native-harness registry
(designs/harness-modular-registry-proposal.md). Introduces the behavior
side-channel that later PRs will dispatch through; no hub is rewired yet, so
this changes no runtime behavior.
- Add `NativeHarnessProvider` (frozen dataclass of dotted import-path strings
for a native harness's lifecycle hooks) and the `native_providers` field on
`HarnessContribution`, plus `native_providers()` / `native_provider_for_key()`
accessors.
- Populate 11 built-in provider rows uniformly from the `omnigent.<key>_native`
module layout (`run_<key>_native`, `_materialize_<key>_agent_spec`, and the
`_auto_create_<key>_terminal` builder re-exported from `omnigent.runner.native`).
Hooks that are still runner closures / inline dispatch (interrupt, stop,
spawn-env, bridge-dir) stay None until those hubs migrate onto the seam.
- Add `omnigent/native_dispatch.py`: a lazy, per-path-cached resolver over the
existing `load_object`, with `resolve` / `resolve_hook` / `resolve_hook_for_key`
so hubs resolve a hook instead of branching on `key == "<x>"`. Import hygiene
preserved — provider rows hold strings; only the resolver imports the target
modules, and only at dispatch time.
- Tests: provider rows cover every native agent 1:1, required hooks are set, and
every populated built-in path actually resolves to a callable (guards against
a typo'd path or renamed symbol); resolver colon/dot forms, caching, and
unset-hook / unknown-key None paths.
The validator still rejects community native metadata (Phase 2 flips it).
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* docs(harness): add implementation-progress ledger (PR 1.1)
Add an append-only "Implementation progress" ledger to the modular-registry
proposal so each PR in the stack records its own status without editing the
plan tables (which would conflict across the 1.1→1.2→1.3 stack on every
rebase). Seed it with 1.1 (#3239, in review).
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* 🔧 chore(lint): Block hardcoded model pins
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* 🔧 chore(lint): Tighten model baseline guard
- Reject duplicate path/model rows so baseline allowances cannot silently accumulate.
- Document heuristic false-negative and multiline-config gaps, plus the bounded full-scan tradeoff.
- Add focused coverage for duplicate baseline validation.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* 🔧 chore(lint): Guard model scan configuration
- Cross-check the pre-commit trigger against the scanner's tracked roots, extensions, exclusions, and allowlist path to prevent silent drift.
- Share the source-extension set across path discovery and scanning.
- Report malformed allowlist counts with consistent path and line context; cover both review cases with focused tests.
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(ui): sub-agent sessions never show reconnect modal when runner dies
A sub-agent session with a dead runner classified as local_stranded,
which disabled the composer and showed the CLI reconnect modal — a
flow designed for top-level host-bound sessions. Sub-agents have no
host binding and can't be relaunched from a CLI command; they recover
via their parent's live runner (server-side heal, #3151).
- Add kind field ("default" | "sub_agent") to Session type and
map it from the wire in sessionFromWire
- Thread kind through LivenessRow and livenessRowFromSession
- Add row 7a in useSessionLiveness: sub_agent with dead runner →
runner_asleep (composer open) instead of local_stranded
Fixes#3413
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
# Conflicts:
# web/src/hooks/useSessionLiveness.ts
* fixup: add kind and backgroundTaskCount to sessionsApi test fixture
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* test(e2e_ui): sub-agent dead runner keeps composer open, no reconnect modal
Regression test for #3413: a sub-agent session with a dead runner was
classified as local_stranded, showing the CLI reconnect modal and
disabling the composer. After the fix (kind=="sub_agent" → runner_asleep)
the composer stays enabled and the "Agent disconnected" banner is absent.
Creates a real child session (parent_session_id set → kind="sub_agent"),
patches the browser's health poll to report runner offline, and asserts
the composer is usable and no reconnect banner appears.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fixup: splice kind from session snapshot into livenessRow when sidebar conv present
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fixup: expose kind in SessionResponse so the UI can detect sub_agent sessions
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fixup: don't re-initialize session on heal — parent runner already hosts the child
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fixup: re-init session for native sub-agents, skip for SDK sub-agents
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fixup: regenerate openapi.json for kind field in SessionResponse
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fixup: update heal docstring + add SDK sub-agent no-init test
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* 🐛 fix(server): heal sub-agent stale runner_id on message-send
A sub-agent copies its parent's runner_id at creation and is never
repointed when the parent's runner is relaunched. The message-send path
returned a permanent 503 for any sub-agent whose runner had
idle-timed-out, even while the parent's replacement runner was healthy
(host_id is None short-circuits all existing relaunch paths).
- Extract _heal_subagent_runner_binding_via_parent from
_recover_subagent_status_forward_via_parent: walks the ancestor chain
(immediate parent → root), waits for the live runner tunnel, calls
replace_runner_id on the child, returns the live client
- Wire the heal into the message-send path after the managed-launch
rendezvous, guarded to kind=="sub_agent"; sets
_runner_needs_session_init=True so the child's harness is initialized
on the healed runner before dispatch
- Refactor _recover_subagent_status_forward_via_parent to delegate
binding repair to the shared helper (no behavior change for the
status-forward path)
- Add regression tests: heal succeeds, no-live-ancestor preserves 503,
top-level sessions not treated as recoverable children
Fixes#3067
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
# Conflicts:
# omnigent/server/routes/sessions.py
* fixup: rebase onto main, apply heal to routes_events.py, fix lint
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fixup: fix test payload format and monkeypatch targets for routes_events
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
The AI-agent workflows install the Claude Code / Codex CLIs with a bare
`npm install` after `cd`-ing into a workspace subdir (`.cc-cli` / `.codex-cli`)
that has no package.json of its own. npm then walks up to the nearest ancestor
package.json to resolve the project root.
Once a repo-root package.json was added, that ancestor became the repo root, so
the install landed in `${GITHUB_WORKSPACE}/node_modules` instead of the subdir.
The follow-up `node node_modules/@anthropic-ai/claude-code/install.cjs` (run from
the empty subdir) then failed with MODULE_NOT_FOUND, breaking Polly review,
issue/security triage, doc-sync, and the run-omnigent-agent action. The
`added 2 packages` line (claude-code has zero deps) was the tell that npm had
reconciled the root tree rather than an isolated install.
Install into `${RUNNER_TEMP}/omnigent-{cc,codex}-cli` instead — outside the
checked-out tree, so no ancestor package.json can ever capture the install. This
matches the pattern e2e-ui.yml and flake-stress-ui.yml already use.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* ci: bump @anthropic-ai/claude-code 2.1.170→2.1.212
v2.1.170 has a corrupted npm cache entry on GitHub Actions runners
causing install.cjs to be missing after `npm install`. Bumping to the
current stable (2.1.212) forces a fresh fetch and clears the bad entry.
Also bumps the ci-deps/package.json pin (was 2.1.163) and the
run-omnigent-agent action default to keep everything consistent.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* ci: update pnpm-lock.yaml for claude-code 2.1.212
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
Codex materializes its versioned plugin store (openai-curated templates,
browser, presentations, ...) into $CODEX_HOME/plugins/cache on session
start. Because codex-native points CODEX_HOME at a private per-session
home, codex re-materializes ~44 MB of identical plugin data into every
session — the dominant on-disk cost once the upstream logs_2.sqlite TRACE
bloat (openai/codex#28224) is fixed in codex >= 0.142.0.
Symlink plugins/cache from the shared source home into each private home,
mirroring the existing skills-symlink pattern. The cache is content-
addressed read-only reference data (verified byte-identical to the shared
copy), so unlike config.toml it needs no per-session isolation. Skipped in
minimal (title-sidecar) mode, which runs no plugins. Best-effort: a symlink
failure logs and lets codex repopulate its own copy.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
- Launch bare agy for Google OAuth and verify with agy models\n- Keep Gemini API-key setup available alongside native sign-in
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
Running a full workspace install without filters complained about ignored
build scripts for @anthropic-ai/claude-code, @google/genai, and protobufjs.
These come from the .github/ci-deps package and are legitimate; approving
them lets Scope: all 4 workspace projects
Already up to date
Done in 194ms using pnpm v11.15.1 / undefined
[ERR_PNPM_RECURSIVE_EXEC_FIRST_FAIL] Command "dev" not found at the workspace root run scripts
instead of erroring.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
* fix(claude-native): never launch a bare family alias a gateway rejects
A family alias (opus/sonnet/haiku/fable) selected on a provider config
whose tier has no ANTHROPIC_DEFAULT_*_MODEL pin is canonicalized by
Claude Code to an Anthropic id (e.g. claude-opus-4-8) that gateways
404, failing session start with "There's an issue with the selected
model". Resolve unpinned aliases to the provider's default model in
resolve_claude_native_model_selection, which launch, sticky handoff,
and /model injection all route through.
Also stop offering the static subscription alias rows to provider
configs with no pins: the picker now lists the one model the config is
known to route.
Co-authored-by: Isaac
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
* refactor: trim the unpinned-alias fix to its minimal form
Shorten the resolver docstring and the pin-less catalog fallback, drop
e2e assertions already implied by the single-row count, and fold the
three alias-passthrough regression tests into one.
Co-authored-by: Isaac
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
* fix(claude-native): scope alias remap to endpoints that reject canonical ids
Review feedback on the unpinned-alias guard:
- Only rewrite an unpinned family alias when the config routes through a
gateway/Bedrock endpoint; the Anthropic API (api.anthropic.com or no
endpoint override) resolves aliases natively, so API-key providers keep
their alias routing and the static picker catalog.
- Respect managed-settings tier pins: Claude Code applies them to the
spawned process, so a managed pin means the alias still routes.
- The runner's /model handler now resolves the session launch config
instead of reading the in-memory cache, so alias resolution survives a
runner restart (cold cache previously skipped the remap).
Co-authored-by: Isaac
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
---------
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
- Add .github/ci-deps to the root pnpm workspace so it uses the shared
pnpm lockfile and install machinery.
- Regenerate pnpm-lock.yaml entries for the e2e-ci-deps package.
- Replace npm install --ignore-scripts in ci.yml and flake-stress-e2e.yml with
pnpm install --frozen-lockfile --ignore-scripts --filter e2e-ci-deps.
- Update electron-build.yml to use setup-pnpm and filter installs for web and
web/electron.
- Update omnidev source so the local dev supervisor installs and runs Vite
with pnpm.
- Update developer docs (README.md, CONTRIBUTING.md, web/README.md,
web/electron/README.md, dev/omnidev/README.md, tests/e2e_ui visual/README.md
and COVERAGE_GAPS.md) to reference pnpm commands.
- Add a minimal root package.json with packageManager: pnpm@11.15.1 and remove
the explicit version from .github/actions/setup-pnpm so CI uses the same
source of truth.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
* fix(onboarding): detect agy settings.json as login fallback on macOS
On macOS, agy 1.1.7+ stores OAuth credentials in Keychain and writes
only ~/.gemini/antigravity-cli/settings.json (no oauth_creds.json).
The existing gemini_auth_has_credential() missed this and falsely
reported 'harness antigravity-native is not configured'.
Accept the existence of settings.json as a fallback signal when no
token files are found. This is safe because the caller
(resolve_native_antigravity_launch) uses it only for an informational
warning — agy always re-drives OAuth on first run regardless.
- Update gemini_auth_has_credential() with settings.json fallback
- Update docstrings to document the third detection path
- Update warning message in antigravity_native_launch.py
- Add unit test for settings.json-only detection
- Fix _GEMINI_DIR isolation in existing test
Signed-off-by: ElliotSun <elros1109@gmail.com>
* fix(onboarding): prove agy login via CLI, not settings.json existence
The macOS lockout this fixes is real: agy 1.1.7+ keeps OAuth in the
Keychain and writes no token file, so the file-only check reported
antigravity-native as unconfigured and connect.py refused to spawn a
runner for a user who was in fact signed in.
Accepting the bare existence of ~/.gemini/antigravity-cli/settings.json
as the fallback signal does not work, because omnigent creates that file
itself: the CLI launch path calls ensure_agy_feedback_survey_disabled
under the real home before agy starts, and build_agy_launch emits no HOME
override. One `omni antigravity` run therefore satisfied the credential
gate forever, on every platform — turning a hard launch gate into a
no-op and letting a runner spawn that dies on its first turn. That is
worst on headless hosts, where agy's OAuth prompt has no TTY.
Ask the CLI instead. `agy models` exits 0 only when signed in and reads
the credential wherever agy stored it, Keychain included, so nothing
omnigent writes can satisfy it. This mirrors ambient._claude_login_detected,
which already solves the identical Keychain split for Claude Code, and
reuses the probe harness_install already wires as the gemini family's
status command.
The fallback is gated on macOS: Linux writes a real token file, so its
absence is a true negative there and the fallback would only add a
subprocess while weakening a signal that works. Failures — missing
binary, non-zero exit, timeout, unreadable home — all read as False,
because readiness must never raise.
Content inspection of settings.json was the alternative considered. It
was rejected as unverifiable from here: no key in that file is known to
mark a completed sign-in on 1.1.7, so keying on one risks reintroducing
the very lockout being fixed.
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
* docs(skills): note agy's macOS Keychain credential in the e2e pre-flight
The pre-flight tells the reader agy's token lives under ~/.gemini, which
leaves a Mac developer on agy 1.1.7+ hunting for a file that is never
written. Name the Keychain case and the `agy models` fallback that
gemini_login_detected() now uses there.
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
---------
Signed-off-by: ElliotSun <elros1109@gmail.com>
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
Co-authored-by: SabhyaC26 <sabhyachhabria@gmail.com>
* fix(cursor-native): auto-accept lingering tool gates under --yolo
cursor-agent's Run Everything mode still sometimes leaves pendingToolCall
markers long enough for Omnigent to mirror ApprovalCards and stall a
piloted parent. When the session launched with --yolo/--force/-f, accept
those tool gates in-pane instead of parking a web card; AskQuestion still
surfaces as deliberate human input.
Co-authored-by: Cursor <cursoragent@cursor.com>
Signed-off-by: Sam Armstrong <armstrongflg@gmail.com>
* fix(cursor-native): satisfy ruff format and PIE810 on yolo args
Co-authored-by: Cursor <cursoragent@cursor.com>
Signed-off-by: Sam Armstrong <armstrongflg@gmail.com>
* fix(cursor-native): make yolo auto-accept bounded and fail-closed
Auto-answering a tool-approval gate is a safety boundary, so the accept path
now refuses to act on anything it cannot confirm, and always has a way out.
The accept was previously a blind keystroke loop: it never checked that a
prompt was on screen, recorded a send to a dead pane as a success, and had no
attempt cap or fallback. A gate that `y` does not clear therefore degraded from
a visible stall into a literal `y` typed into cursor's composer every two
seconds for the life of the session, with no card ever surfaced.
The accept key now goes out only while `capture_cursor_pane` shows cursor's
parenthesised accept hint, at most three times, and at most once per poll pass
(cursor renders one prompt at a time). A dead pane, a send tmux rejects, or a
gate still pending after the budget all fall back to the same ApprovalCard the
non-yolo path shows, so the worst case is the visible stall we have today.
Because a call accepted this way is never seen by a human, the INFO line now
carries an argument preview: it is the only record Omnigent approved the call.
`cursor_launch_args_enable_yolo` was failing open in the same spirit —
`--yolo=false` and `--force=false` both read as enabled, because only the
presence of the `=` form was checked. Explicit off-values are now honoured, and
a bare `--` ends the flag scan so a `-f` in the prompt text that follows is
text rather than a request to bypass approvals.
Tests cover the bounded retry, the fallback to a card, an idle pane, a dead
pane, an undelivered keystroke, an explicit non-yolo session, and the
off-value / end-of-flags argv cases. The design doc gains a section on the
fail-closed contract and drops its claim that Omnigent never sends a verdict of
its own initiative; its stale `Code:` pointer at the runner wiring is refreshed
to where that wiring now lives.
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
* fix(cursor-native): re-apply yolo wiring where auto-create now lives
`_auto_create_cursor_terminal` moved out of `omnigent/runner/app.py` into
`omnigent/runner/native/orchestration.py`, which left `app.py` a re-export
shell and this branch's wiring hunk applying to code that no longer runs.
Derive `auto_accept_approvals` from `launch_config.terminal_launch_args` at the
live call site instead.
This kwarg is the only thing that turns the in-pane auto-accept on, and it is
one line inside a large function, so a future move can drop it and leave the
feature inert with the whole suite green. Pin it: the auto-create harness now
captures the elicitation supervisor's kwargs, and a parametrized test asserts
the derived stance for `--yolo`, `--force`, `-f`, `--yolo=false`,
`--auto-review`, and no args. Deleting the kwarg fails all six.
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
---------
Signed-off-by: Sam Armstrong <armstrongflg@gmail.com>
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: SabhyaC26 <sabhyachhabria@gmail.com>
The runDebug, listDevices, and reverseProxy Exec tasks called
`commandLine("adb", ...)`, relying on adb being on PATH. The Gradle
daemon is long-lived and may have been started from an environment
whose PATH doesn't include platform-tools (e.g. homebrew's
android-commandlinetools), so the spawn fails with
"A problem occurred starting process 'command 'adb''" — even though
AGP's own installDebug succeeds because it resolves adb from the
SDK directory internally.
Resolve adb from android.sdkDirectory instead, mirroring AGP, so
the custom launch tasks are independent of the daemon's PATH.
* docs(DBSPEC): remove stale DBOS/tasks references
The tasks table and DBOS were removed (migration
b9c1d2e3f4a5_drop_tasks_table), but DBSPEC.md still described the
old DBOS-backed workflow design: the tasks table schema, the
try_deliver/close_inbox steering handshake, and the TaskStore
method mapping. Updated the doc to match current state — turn
state now lives in-memory in the runner (_active_turns,
_session_message_buffers), and conversation_items.response_id is
just an app-generated grouping id with no backing table.
Also added the created_by column to conversation_items, which
existed in code but was missing from the doc.
Signed-off-by: Harry Su <tiecheng.su@robinhood.com>
* docs(DBSPEC): correct FK section — no DB-enforced FKs, cleanup is explicit app code
Addresses the blocking review: the previous revision claimed an ON DELETE
CASCADE FK on conversation_items.conversation_id, but
p1a2b3c4d5e6_remove_all_fks dropped every FK (Rule R032) and
delete_conversation cleans up children before parent explicitly. Also
precision-fix response_id as harness- or app-generated per review.
Signed-off-by: Harry Su <tiecheng.su@robinhood.com>
* docs(DBSPEC): correct table count, deletion order, and position allocator
The accuracy pass left five claims that don't match the code:
- The opening line said four tables in the default schema. There are 17
in `db_models.py`, and none sets an explicit schema — the same doc names
labels, comments, and policies as tables a hundred lines later. Scope the
sentence to the four tables this doc covers and point at the models as the
full list.
- `delete_conversation` was described as deleting comments and policies
before the conversation rows. It uses two transactions: the AP one drops
FTS rows, items, labels, and the conversation rows; a second best-effort
transaction then cleans up comments, policies, session permissions,
conversation metadata, and session-scoped agents *after* the conversation
is gone. The doc also omitted three of those tables and hid the
best-effort tradeoff the method's own docstring calls out.
- "Turn state is not persisted to this schema at all" was overstated. The
authoritative state is in-memory, but `persist_live_status` mirrors
`live_status` / `pending_elicitation_count` onto
`omnigent_conversation_metadata` so any replica can render session status.
- The "Delete agent" row documented cancelling in-flight turns for the
agent's live sessions. No such mechanism exists: `AgentStore.delete` is a
bare row delete with no production caller and no HTTP route, and
session-scoped agent rows are removed by `delete_conversation`.
- The position allocator no longer runs `SELECT MAX(position) + 1`.
`append()` reads and advances the `conversations.next_position` counter
under `_lock_conversation`, keeping allocation O(1); the `MAX(position)`
scan survives only as a one-time backfill for pre-counter conversations.
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
---------
Signed-off-by: Harry Su <tiecheng.su@robinhood.com>
Signed-off-by: SabhyaC26 <sabhyachhabria@gmail.com>
Co-authored-by: SabhyaC26 <sabhyachhabria@gmail.com>
- Define a root pnpm-workspace.yaml with web/ and web/electron/ packages.
- Move npm overrides from web/package.json into workspace overrides, using a
shared catalog: for react, react-dom, and shiki.
- Preserve 7-day dependency cooldown via settings.minimumReleaseAge: 10080.
- Delete web/package-lock.json and web/electron/package-lock.json; add the
generated root pnpm-lock.yaml.
- Update web/electron/package.json scripts to use pnpm --filter web run build:overlay.
- Remove web/.npmrc and web/electron/.npmrc; no committed .npmrc (CI forces the
public registry via env var).
- Add .github/actions/setup-pnpm so all workflows can share a pinned pnpm
11.15.1 + Node setup.
- Convert lint.yml and web-tests.yml to pnpm; update ui-snapshot and e2e-ui
workflows.
- Update the web-prettier pre-commit hook to run web/node_modules/.bin/prettier
directly when present.
- Update justfile to prefer pnpm for Electron recipes and lockfile normalization.
- Ensure remaining npm-based workflows (editors/vscode/, .github/ci-deps/,
deploy/cloudflare/) are untouched and continue to work.
- Add pdfjs-dist worker URL import so Vite emits the worker asset under pnpm's
hoisted node_modules layout.
- Force shiki and its first-party packages into a single build chunk to avoid a
Cyclic top-level import that produced a 'flatMap' runtime error in Monaco.
- Pin build-tool versions to the legacy npm lockfile (vite 8.1.0, tailwindcss
4.3.1, jiti 2.7.0, lightningcss 1.32.0, postcss 8.5.15) so bundler behavior
stays consistent with the pre-migration builds.
- Update tests/e2e_ui/test_pwa_build.py to omit the now-incorrect -- separator
when forwarding --outDir to pnpm run build:embed.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
On a claude-native session with intelligent routing on, the routed model
was selected but the user's first message was silently dropped — the model
switched, no error surfaced, but no turn ran.
The server issued TWO unsynchronized writes to the same tmux pane: a
standalone model_change event (which typed /model <routed> into the pane)
AND, separately, the user's message (typed in via inject_user_message).
These raced. The message keystrokes landed mid-switch, inject_user_message
never saw its draft, hit its submit-blind fallback, and returned without
error. Model applied, message gone.
Fix: remove the second writer by folding the switch into the message turn,
mirroring how the SDK/pi path already applies the routed model as one
operation.
- Executor (ClaudeNativeExecutor.run_turn): the routed model already
arrives in ExecutorConfig.model and was being discarded. It is now
applied: when config.model differs from the pane's model, type /model
then inject the message — both under the existing _inject_lock, in
order, exactly once. inject_user_message's prompt-ready gate + verified
submit then guarantee delivery. _applied_model is seeded lazily from
read_launch_model so turn 1's routed pick is compared against the spawn
model rather than blindly re-issued.
- Server (_sessions/orchestration.py): the routed model rides in-band on
the message (model_override, an extra field the harness MessageEvent
forwards into ExecutorConfig.model), and the separate racing model_change
POST is dropped. The manual composer /model picker path (PATCH ->
model_change) is untouched.
Adds three executor tests: /model precedes the message in order under one
lock; no /model without a routed model; no /model when already on the
routed model. The ordering test fails against the prior discard-config
behavior.
Signed-off-by: Edwin He <41037314+Edwinhe03@users.noreply.github.com>
* Bump version to 0.8.0.dev0
Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* chore(release): keep uv.lock at main's shape, stamp workspace versions only
The bump workflow's full relock rewrites every entry with new-uv metadata
churn; restoring main's lock and stamping just the workspace versions keeps
the PR reviewable. Workspace package blocks verified identical to the
relocked version.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* fix(codex): attribute per-model usage for turns with no pinned model
codex_executor's TurnComplete.usage never carried a "model" field, unlike
every other relay executor (claude-sdk, cursor, copilot, openai-agents,
pi). For a codex-harness agent that pins no llm.model (e.g. Debby's
gpt head, which deliberately defers to the harness/provider default),
_accumulate_session_usage's model-resolution fallback chain had nothing
to resolve to, so the turn's flat token/cost totals still accumulated
but session_usage.by_model silently never got an entry for it.
Stamp the turn's resolved model (already in scope as run_turn's `model`
argument) onto the usage dict extracted from tokenUsage/updated, mirroring
claude_sdk_executor's observed_model pattern.
Signed-off-by: Sato Taiga <antfgxgres@gmail.com>
* test(sessions): add regression test for codex per-model usage attribution
Exercises the real _accumulate_session_usage and GET /v1/sessions/{id}
API against a codex-harness agent with no pinned llm.model (Debby's gpt
head's exact shape): a usage delta with no "model" key still accumulates
the flat total but leaves by_model empty (the bug), while one carrying
"model" (as codex_executor.py now stamps it) gets a by_model entry that
also surfaces through the session snapshot the web UI's cost panel reads.
Signed-off-by: Sato Taiga <antfgxgres@gmail.com>
---------
Signed-off-by: Sato Taiga <antfgxgres@gmail.com>
N/A
- Added `min_version` and `max_version_exclusive` to `HarnessInstallSpec` and made `harness_cli_installed` probe `--version` when bounds are declared, so setup and dispatch fail loud for outdated CLIs.
- Implemented generic `--version` parsing + PEP 440 comparison with date-version normalization so Cursor and Hermes calendar-version strings compare correctly.
- Wired code- and changelog-derived version floors for all CLI-backed native harnesses (e.g. Claude >=2.1.161, Codex >=0.137.0, Cursor >=2026.06.02, Kimi >=1.47.0, Hermes >=2026.06.05).
- Updated the CLI setup overview and install prompt to show "Needs upgrade" and the detected/declared versions instead of claiming a present-but-outdated CLI is "not installed".
- Added the `version-too-low` readiness reason and surfaced it in the web UI badge/notice; also made Cursor native auth-aware so it now reports `needs-auth` when installed but not logged in.
- Fixed the readiness-layer lookup so `version-too-low` correctly surfaces for all native harnesses that declare a version floor (Claude, Cursor, OpenCode, Kiro, etc.) instead of falling back to `binary-missing`.
- Preserved the existing `antigravity-native` credential gate: an installed `agy` CLI without a stored Gemini credential still reports not-ready.
- Added E2E UI coverage for the new `version-too-low` warning and updated readiness unit tests for version-bound and credential-bound behavior.
```bash
uv run pytest tests/onboarding/test_harness_install.py \
tests/onboarding/test_harness_readiness.py \
tests/cli/test_configure_models.py \
tests/test_codex_native.py -q
npm run --silent test -- --run src/lib/harnessSetup.test.ts src/shell/NewChatDialog.test.tsx
```
N/A — the change is mostly backend/UX copy; no new visual components.
- [x] Feature
- [ ] Bug fix
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [x] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
Manual verification: ran targeted backend/web test suites after each change and confirmed `omnigent setup`/`harness_cli_installed` now report “installed (vX) but not supported” rather than “missing” for outdated CLIs.
Native harness setup now checks the installed CLI version and prompts to upgrade if it is too old.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
The MITM egress proxy verifies upstream TLS against the system trust
store built by _system_ca_bundle(). It read only the consolidated
cafile (get_default_verify_paths().cafile/openssl_cafile) and ignored
the capath directory. Corporate MDM / IT-managed roots are commonly
installed as loose files under capath (with hashed symlinks) rather than
merged into the cafile, so they were missing from the proxy's trust
store. Any upstream host whose chain relies on such a root then failed
verification (e.g. a corp-intercepted github.com returned 502 from the
proxy) even though the host's own tools trusted it.
Read capath too: concatenate the loose PEM certs from the capath
directory onto the cafile bundle (dedup by resolved path, skip non-PEM
entries), keeping the certifi fallback when neither yields any certs.
Added tests: a CA present only as a loose capath file lands in the
bundle, and non-PEM files in capath are skipped.
Signed-off-by: Nishith Sinha <308961816+nhsdb@users.noreply.github.com>
Co-authored-by: Nishith Sinha <308961816+nhsdb@users.noreply.github.com>
Tools invoked by generic name (awk, python3, editor, pager, ...) resolve
through /usr/bin/<name> -> /etc/alternatives/<name> -> real binary. The real
binaries already live under the mounted /usr, but /etc/alternatives was not
bound, so the intermediate symlink node was missing inside the jail and the
lookup failed with 'command not found'.
Bind /etc/alternatives read-only in the default _DEFAULT_ETC_DIRS list,
alongside the existing /etc/ssl and /etc/ca-certificates dir binds. It is a
directory of symlinks (no secrets); read-only means the mapping cannot be
repointed, and every target is a binary already exposed under /usr, so this
grants no new capability -- it only restores standard name resolution.
Linux (bwrap) backend only; darwin_seatbelt is unaffected by this mechanism.
Signed-off-by: Nishith Sinha <308961816+nhsdb@users.noreply.github.com>
Co-authored-by: Nishith Sinha <308961816+nhsdb@users.noreply.github.com>
Native harnesses (claude-native / codex-native) report a cumulative
SESSION total, not a per-model split. `_persist_native_cumulative_usage`
SET each active model's `by_model` bucket to the whole running total, so a
session that switched models mid-run double-counted the shared baseline:
the previous model kept its last cumulative snapshot while the new model
was set to the full total, and summing the buckets exceeded the session
total (e.g. total $11.91 but opus $10.80 + sonnet $11.91).
Attribute only each report's growth (new - old) to the currently-active
model instead, mirroring the relay path's per-model delta accumulation.
Per-model token and cost buckets now hold each model's own usage and sum
to the flat session total across model switches. Deltas are clamped >= 0
so a lowered / rebased report never claws usage back out of a bucket (the
flat totals are likewise monotonic-clamped).
Read-only reporting (`omni usage`, the web session sidebar) needs no
change — it reads `by_model` verbatim, so corrected data flows through.
Existing sessions keep their already-stored buckets; this corrects
attribution for turns recorded after it ships (not backfillable).
Co-authored-by: Isaac
Signed-off-by: Jakub Majorek <majorek.jakub@gmail.com>
* feat(web): add a harness credential from the New Chat setup dialog (M3 frontend)
Frontend for Setup From the Web UI — turn a yellow needs-setup harness
green from the browser (Claude/Codex/Pi) via an inline equal-weight auth
form (adopt / subscription signpost / API key / gateway), plus the setup
dialog UX cleanups. Gated behind the existing harness_install_enabled cap.
Rebased onto latest main (the M3 backend #3088 is now upstream, so only
web/ + follow-up backend fixes remain) and folded in the Polly review
notes: stable option keys, clear secret fields on save, and a note that
default_model/wire_api are backend-accepted but reserved for a follow-up.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(web): scope useHosts refocus-refetch to the setup flow (Polly review)
staleTime:0 + refetchOnWindowFocus was app-wide across ~8 useHosts
consumers, bumping /v1/hosts volume on every refocus. Make it an opt-in
refetchOnFocus flag; only the setup dialogs (NewChatDialog, HarnessSetupDialog)
that need live readiness recovery pass it. Others keep the 30s stale window.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(web): guard the credential form against double-submit + close test gaps
Address Pat's review:
- Gate both form onSubmit handlers on !busy so hitting Enter in the field
during an in-flight save can't re-POST the secret (the Save button was
already disabled, but the keyboard path wasn't guarded).
- Add a double-submit-guard test, plus direct hook tests for
useStoreCredential (path/body split, JSON detail + non-JSON error parse,
cache patch + detect invalidation) and useDetectedCredentials
(GET/parse, empty-body fallback, enabled/host gating).
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(web): let Pi adopt an openai-family credential too (Polly review)
Pi consumes both anthropic and openai and the daemon adopts a detected
credential under its OWN family, so a host with only $OPENAI_API_KEY could
back Pi — but the adopt filter scoped to Pi's single write-default family
(anthropic), hiding that affordance. Add harnessCredentialAdoptFamilies
(Pi -> both families) and filter the adopt row on it; the paste/gateway
paths and the cross-family guard for Claude/Codex are unchanged.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
---------
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(codex-native): carry hook trust across private CODEX_HOME copy
When codex-native provisions a per-session private CODEX_HOME and copies
config.toml into it, the [hooks.state] keys inside the copy still reference
the global ~/.codex/ paths. Codex keys trust records by the absolute path of
the hooks file, so every key misses and Codex opens an interactive "Hooks need
review" prompt on every launch. Headless sub-agents can never answer it, so
the app-server never emits thread/started and the run dies on the 15s timeout.
Fix: two changes to _populate_codex_home_config:
1. Symlink hooks.json from the global home into the private home (alongside
auth.json). This makes the user's hooks reachable at the private path.
2. After copying config.toml, rewrite [hooks.state.*] key path prefixes from
source_dir to target_dir. The hash values are left untouched, so trust is
neither widened nor weakened — it is only carried across the copy that
Omnigent itself performs.
Fixes#3268.
Signed-off-by: Tomu Hirata <tomu@omnigent.ai>
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: gate hooks.json symlink on not minimal_config; drop redundant re import
The minimal_config path rebuilds config.toml from scratch with only
model_provider/model_providers/profiles — no [hooks.state] entries.
Symlinking hooks.json there with no trust state re-introduces the
interactive trust prompt for the title worker. Gate the symlink (and
the trust-key rewrite that gives it meaning) on not minimal_config.
Also remove the redundant `import re as _re` inside
_retarget_codex_hook_trust_keys; re is already imported at module level.
Addresses Polly review feedback on #3343.
Signed-off-by: Tomu Hirata <tomu@omnigent.ai>
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(codex-native): flush accepted hook trust back to global config on close
When a user accepts the hook-trust prompt inside a session, Codex writes
[hooks.state] entries into the per-session private config.toml — but those
are discarded when the session ends because the private CODEX_HOME is
ephemeral. So the prompt reappears on every launch.
Fix: in CodexNativeAppServer.close(), call _merge_codex_hook_trust_back to
read [hooks.state] from the private config.toml, translate the path keys
from the private home back to the global ~/.codex/ prefix, and upsert them
into ~/.codex/config.toml atomically. The next session's _populate_codex_home_config
copies the global config (now with the trust entries), and
_retarget_codex_hook_trust_keys translates the paths forward to the new
private home — so Codex sees the hooks as already trusted and skips the prompt.
The write is best-effort: any failure is logged as a warning rather than
raised, since the session has already ended.
Fixes#3268.
Signed-off-by: Tomu Hirata <tomu@omnigent.ai>
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: assign tmp before try block to avoid unbound variable warning
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu@omnigent.ai>
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(claude-sdk): surface ResultMessage is_error as ExecutorError, not assistant text
When the Claude SDK reports a harness-level failure (e.g. an expired
login or unauthenticated session), the terminal ResultMessage carries
is_error=True and the failure text in result. The executor was ignoring
is_error and assigning result directly to response_text, so the error
appeared in the conversation as though the model had said it — with no
error item, no harness attribution, and no log line.
Fix: check is_error before touching response_text. When true, set
terminal_error (the existing path that yields ExecutorError and returns)
and log an error line naming the agent. When false, the existing
response_text assignment runs unchanged.
Also add is_error to _ResultMessageObj so the Protocol matches the
SDK's actual shape (it was only declared on _ToolResultBlockObj before).
Closes#3282
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(claude-sdk): use getattr for is_error, handle null result, add unit test
Address Polly review feedback on #3342:
- Use getattr(result_msg, 'is_error', None) instead of direct attribute
access so that existing test doubles that only set session_id/result
don't raise AttributeError (matching the sibling getattr calls for
session_id and usage in the same block).
- When is_error=True but result is None/empty, fall back to a generic
'claude-sdk harness error' message rather than silently dropping the
failure.
- Add test_result_message_is_error_yields_executor_error: verifies that
a ResultMessage with is_error=True is routed to ExecutorError and does
not appear in TurnComplete.response.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(test): wrap long assertion string to satisfy ruff E501
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(web): stop short links collapsing table columns in chat markdown
Streamdown styles links with `wrap-anywhere` (overflow-wrap: anywhere),
which also drops the element's min-content width to a single character.
Inside its `table-layout: auto` table that let a link-only column be
squeezed to ~2ch, so a short link like "#3090" stacked one or two
characters per line while the prose columns took all the width.
Narrow links inside table cells to `break-word`: overlong URLs still
soft-wrap, but min-content stays at the longest unbreakable run so the
column can no longer be squeezed below it. Prose links keep `anywhere`.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* test(e2e-ui): guard markdown table link column width in the browser
The CSS fix for the collapsing "PR #" column is only observable with a
layout engine, so the vitest companion can pin the rule and its selector
scoping but not the width. This adds the browser-side half: a seeded
assistant message renders the table shape that triggered the bug — a
link-only `#` column, wide prose columns, and a full-URL column — and
asserts the short link stays on one line box, its cell is at least as
wide as the link, and a long URL still soft-wraps inside its cell.
Verified against the pre-fix stylesheet: `#3090` stacks across 5 line
boxes without the `overflow-wrap: break-word` narrowing, 1 with it.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* fix(seatbelt): resolve two-hop proxy symlink when detecting CPython install root
When a Python interpreter is installed via `uv tool install`, the
executable is a two-layer symlink:
~/.local/share/uv/tools/<pkg>/bin/python → (proxy)
~/.local/share/uv/python/cpython-3.12.X-.../bin/python3.12
The literal proxy path grandparent (`tools/<pkg>/`) has no CPython
`lib/python*` markers, so `_interpreter_install_root` returned None.
`_add_topmost` then raised OSError before ever checking the resolved
path, causing every session to fail with:
darwin_seatbelt: helper interpreter at '.../uv/tools/omnigent/bin/python'
resolves under the unsafe ancestor '/Users'; ...
Fix: in `_add_topmost`, when the literal path yields no install root,
resolve it one level and retry `_interpreter_install_root` on the
resolved path before giving up. The resolved CPython install root
(which does carry the canonical markers) is then granted as the narrow
subpath, matching the existing behaviour for direct uv-python installs.
Also update the OSError message to say 'CPython install root' and note
that both the literal and resolved path were tried, and fix the
matching assertion in the existing test.
Fixes#3237.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(seatbelt): grant pi_dir and $TMPDIR write root so sandboxed pi can boot
Two follow-up fixes found by running `omnigent run --harness pi` with
darwin_seatbelt enabled end-to-end:
1. with_additional_read_roots silently dropped pi_dir
When the spec declares no read_paths, resolve_sandbox returns
read_roots=None (meaning 'no spec-supplied grants').
with_additional_read_roots bailed early on None, so the pi node_modules
dir granted by _try_sandbox_pi was never added to the policy. Result:
pi failed with 'Cannot find package .../pi-ai/index.js' because the
seatbelt profile had no subpath rule for the nvm install tree.
Fix: treat None as an empty list rather than 'already unrestricted' —
the caller is explicitly widening the policy and must be honoured even
when the spec has no grants of its own.
2. PI_CODING_AGENT_DIR was created under $TMPDIR, which wasn't granted
_try_sandbox_pi granted /tmp as a write root, but on macOS $TMPDIR is
/var/folders/.../T/ (not /tmp). PI_CODING_AGENT_DIR is created with
tempfile.mkdtemp() which uses $TMPDIR, so pi got EPERM trying to write
its extension/settings. Fix: also grant tempfile.gettempdir() alongside
/tmp.
With all three fixes (two-hop symlink detection, read-roots None handling,
TMPDIR grant) `omnigent run /tmp/pi-sandbox-bundle --harness pi` boots and
completes a full turn end-to-end under darwin_seatbelt.
Fixes#3237.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: route kimi and inkling through Responses API via system.ai.* ids
Kimi and inkling never send finish_reason in /chat/completions streaming
responses, causing Pi to throw 'Stream ended without finish_reason'.
These models work correctly via the Responses API at /ai-gateway/codex/v1
using their system.ai.* model ids (system.ai.kimi-k2-7-code,
system.ai.inkling).
- Add system.ai.kimi-k2-7-code and system.ai.inkling to
_DATABRICKS_RESPONSES_MODELS in the executor
- Add _DATABRICKS_TO_SYSTEM_AI mapping in pi_native_credentials so live
endpoint fetch translates databricks-* ids to system.ai.* and routes
them to the gpt_responses bucket (openai-responses at /ai-gateway/codex/v1)
- Update _pi_needs_responses_api to treat system.ai.* models as responses
- Update _pi_provider_for_model to route system.ai.* to databricks-openai
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(review): restore substring reasoning fallback and fix run-path translation
Addresses Polly's review of #3307:
1. Restore 'kimi'/'inkling' to substring reasoning check in _fetch_pi_model_lists
so unmapped variants (renamed/versioned endpoints not in _DATABRICKS_TO_SYSTEM_AI)
still get reasoning:true — preventing silent regression.
2. Translate databricks-* model ids to system.ai.* in the executor run path
(_build_env_and_dir) so model_override='databricks-kimi-k2-7-code' correctly
routes to the databricks-openai (Responses API) provider, not databricks-completions.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: move GLM to Responses API via system.ai.glm-5-2
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: move Qwen3 to Responses API via system.ai.* ids
Qwen3 returns array content with tool calls via /chat/completions causing
[object Object] errors. system.ai.qwen3-next-80b-a3b-instruct and
system.ai.qwen35-122b-a10b work correctly via the Responses API.
Also removes qwen3 from _unsupported_in_pi since it's now handled.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor: replace hardcoded system.ai map with keyword-based detection
- Replace _DATABRICKS_TO_SYSTEM_AI exact-id dict with _databricks_to_system_ai()
function that detects by keyword (kimi, inkling, glm-5, qwen3, qwen35) and
derives system.ai.* id by stripping 'databricks-' prefix. Handles future model
variants automatically without needing to update an exact-id map.
- Apply the same swap in model_catalog._fetch_databricks_listing so sys_list_models
returns system.ai.* ids directly, letting the LLM use the correct id immediately.
- Use specific fragments (glm-5 not glm) to avoid false-positives like
zai-org-glm-4-7 which has no system.ai.* alias.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(review): fix _ensure_rpc selector translation; revert GLM to completions path
Addresses Polly's blocking issues:
1. Normalize model id to system.ai.* at the top of _ensure_rpc so that both
models.json and the provider/model selector see the same id. Previously only
_build_env_and_dir translated the id but _ensure_rpc still built the selector
from the untranslated databricks-* id, causing 'Model not found' in Pi.
2. Revert GLM (databricks-glm-5-2) back to the completions path. GLM works fine
via /chat/completions with finish_reason=true — moving it to the Responses API
was unnecessary and undocumented. Removed from _SYSTEM_AI_MODEL_KEYWORDS and
_DATABRICKS_RESPONSES_MODELS.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor: use Unity Catalog model-services API for Pi model discovery
Replace /api/2.0/serving-endpoints with /api/2.1/unity-catalog/model-services
which returns system.ai.* model ids directly with supported_api_types metadata.
Benefits:
- No databricks-* → system.ai.* translation needed
- Authoritative API capability info: models with 'openai/v1/responses' in
supported_api_types go to the Responses provider; others to completions
- Embeddings excluded cleanly via has_embedding check
- sys_list_models returns system.ai.* ids directly via _fetch_databricks_uc_listing
Also add _ensure_rpc id normalization so databricks-* model_override values
are translated before building the provider/model selector.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: route all system.ai.* models through AI Gateway (omnigent-openai)
system.ai.* ids are not valid at /serving-endpoints — they only work
via the AI Gateway at /ai-gateway/codex/v1. Previously, system.ai.*
models without openai/v1/responses in UC metadata (kimi, inkling,
qwen3) were routed to omnigent-completions at /serving-endpoints,
causing 404 errors.
Route all system.ai.* models to omnigent-openai regardless of UC
supported_api_types.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(test): update test to expect all system.ai.* models in gpt_responses
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi-native): surface Pi model errors as visible error items in web UI
When Pi's API call fails (e.g. 404 for unknown model id, 400 for
unsupported API type), the extension was silently returning from
message_end with no output, leaving users with an empty turn.
Post an external_conversation_item of type 'error' when message.stopReason
is 'error', so the error appears in the web UI chat.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(tests): update model_catalog tests for Unity Catalog API format
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: revert Qwen3 from responses API - Pi sends fields that Qwen3 rejects
/ai-gateway/codex/v1/responses rejects Pi's standard Responses API fields
(parallel_tool_calls, temperature:null, top_p:null) for Qwen3, causing 400.
Route Qwen3 back to omnigent-completions until either:
- Pi adds compat flags to suppress these fields for non-standard providers
- The upstream array-content fix (earendil-works/pi#7062) lands to fix [object Object]
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: restore Qwen3 to Responses API path via system.ai.*
Pi only sends store:false in requests - the earlier 400 was from a stale
session before the routing fix. Confirmed minimal Pi request works fine
for Qwen3 via /ai-gateway/codex/v1/responses.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(review): scope UC listing to pi path only; fix test fixtures
Polly's review correctly identified that using _fetch_databricks_uc_listing
for all Databricks providers leaks system.ai.* ids to non-pi harnesses
(claude-sdk, codex, openai-agents) that only understand databricks-* ids.
Revert model_catalog.py to use _fetch_databricks_listing (serving-endpoints)
for sys_list_models. _fetch_databricks_uc_listing remains available but is
only used internally by pi_native_credentials._fetch_pi_model_lists.
Also fix test_model_catalog.py fixtures to use the correct serving-endpoints
payload shape (databricks-* ids) rather than the UC model-services shape
(system.ai.* ids) which the non-pi listing never emits.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* test(model_catalog): update pi tests for UC model-services API
Pi harnesses now call `/api/2.1/unity-catalog/model-services` and return
`system.ai.*` model ids instead of `databricks-*` ids. Update the test
fixtures and expected ids to match:
- `_databricks_transport`: now serves both the serving-endpoints page
(non-pi) and a UC model-services page (pi harness calls).
- `test_databricks_listing_filters_to_chat_llms`: expect `system.ai.*`
ids and matching family assertions.
- `test_databricks_listing_skips_explicitly_non_ready_endpoints`: rewrite
to use UC format (UC has no per-service readiness flag).
- `test_listing_failure_reported_and_not_cached`: switch to codex-native
harness to test generic failure/retry without UC routing complexity.
- `pi-everything` parametrize: update expected ids to `system.ai.*`.
- `model_catalog.py`: add TTL cache for UC listings (same `_listing_cache`
with a `"uc:"` prefixed key) so pi harness calls cache-hit correctly.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(model_catalog): fix ruff RUF005 and E501 lint errors
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(test_model_catalog): shorten docstring to fix E501
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi_executor): scope system.ai.* Responses-API routing to kimi/inkling/qwen only
system.ai.claude-* and system.ai.meta-llama-* ids should route to their own
providers (Anthropic surface and completions respectively), not the Responses
API. Previously _pi_needs_responses_api returned True for *all* system.ai.*
ids, which would have routed llama to the Responses endpoint.
Fix: check _SYSTEM_AI_MODEL_KEYWORDS in the system.ai.* branch so only kimi,
inkling, and qwen3 variants return True. Claude is already caught upstream by
the "claude" substring check in _pi_provider_for_model.
Also update stale docstrings in _needs_responses_api and _unsupported_in_pi
that still mentioned qwen3 as excluded (it was re-enabled via the Responses API).
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(pi): route GLM via Responses API (system.ai.* ids)
GLM has the same finish_reason issue as Kimi/inkling on /chat/completions.
Route it through the AI Gateway Responses API by adding "glm-" to
_SYSTEM_AI_MODEL_KEYWORDS (uses "glm-" not bare "glm" to avoid matching
"zai-org-glm-4-7" which has no system.ai.* alias).
- Remove GLM from _PI_REASONING_MODEL_FRAGMENTS (reasoning:true is a
completions-path flag; not needed for Responses API).
- Remove GLM from the reasoning:true assignment in _fetch_pi_model_lists.
- Update test: kimi no longer gets reasoning:true (Responses API path).
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): remove gpt-oss from _unsupported_in_pi; it routes via Responses API
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): exclude all Gemini models from Pi, not just gemini-2-5
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): exclude only gemini-2-5 from Pi; other Gemini models use completions
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): drop redundant qwen35 keyword; qwen3 already matches qwen35 ids
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor(pi): remove _databricks_to_system_ai; catalog always returns system.ai.* for pi
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): remove reasoning:true from kimi/inkling static model entries
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(pi): route Gemini via /ai-gateway/mlflow/v1/chat/completions using system.ai.* ids
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): fix _unsupported_in_pi to only exclude gemini-2-5; gemini-3+ route via mlflow
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor(pi): remove static kimi/inkling/qwen3 entries from _DATABRICKS_RESPONSES_MODELS
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): route system.ai.* llama/other models to mlflow gateway; rename provider to databricks-mlflow
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): use generic base URL for non-Databricks providers (OpenAI API key, LiteLLM)
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi): address Polly review — fix 4-tuple annotation, system.ai.gpt routing, gpt-oss exclusion, UC listing filter
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(model_override): strip system.ai.* prefix for vendor-direct providers (OpenAI key, etc.)
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
A native Pi session routed through a Databricks gateway whose OAuth token
can't be resolved (expired refresh token) launched fine but every message
silently failed to reach the model — no reply, no error. `_databricks_pi_provider`
caught all failures in one try/except and still returned a provider whose
`!databricks auth token` apiKey fails at request time; because pi-native
dispatches turns fire-and-forget, the failure never round-tripped back as an
Omnigent error.
Split credential resolution from the (benign) model-list fetch so a genuine
auth failure carries a `credential_warning`. At terminal auto-create, surface
that warning as an `error` item via `external_conversation_item`: it renders as
the web UI's distinct error banner (not a misleading assistant bubble),
persists across reload, is a non-content item type so it never enters the next
turn's context, and posts without queuing an agent turn (safe on a session
whose model is unreachable).
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Claude Opus 5 (released 2026-07-24) was missing from the curated
_SUBSCRIPTION_STATIC_MODELS["claude"] list. Verified empirically against
Claude Code 2.1.220: 'claude-opus-5' -> is_error:false; the dated form
'claude-opus-5-20260724' and a 'claude-opus-5-fast' variant both return
is_error:true, so neither is added.
Placement follows the existing convention: tiers descend
fable -> opus -> sonnet -> haiku, newest version first within a family
(matching claude-sonnet-5 ahead of claude-sonnet-4-6), so opus-5 slots
between fable-5 and opus-4-8.
The web mirror (web/src/lib/claudeNativeModels.ts) needs no change: it
lists version-agnostic aliases ('opus' resolves to the latest Opus) by
design, not pinned ids.
Signed-off-by: Abdullah Said <abdullahsaid89@gmail.com>
Co-authored-by: omnigent <noreply@omnigent.ai>
* ci(ui-snapshot): make the visual-baseline gate merge-blocking
The UI Snapshot visual-regression check was advisory ([non-blocking]) and
not in the required-checks set, so a UI change could land without
regenerating the committed baselines — which is how the baselines drifted
stale on main (every PR since #3311 fails the gate identically).
Register it as a required merge gate:
- Drop the "[non-blocking]" suffix from the job name.
- Add "UI Snapshot (visual baselines)" to REQUIRED and ALLOW_SKIP in
merge-ready/required.sh, plus a workflow_for mapping. It's safe as a
required check: a PR touching no render input skips the render via the
`detect` job's `if` gate, and an if-skipped job reports success — so
non-UI PRs satisfy the check instead of sitting pending. ALLOW_SKIP +
workflow_for let the gate tell that genuine skip from a still-pending run.
- Add "UI Snapshot" to merge-ready.yml's workflow_run triggers so the gate
re-evaluates when the snapshot workflow completes.
- Update the visual README's merge-blocking section.
This PR edits ui-snapshot.yml (a render input), so the gate runs here and
fails on the stale baselines; the `update-ui-snapshot` label regenerates
them onto this branch to turn it green.
Co-authored-by: Isaac
* test(e2e-ui): regenerate visual baselines
---------
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
* fix(sessions): fall back to localStorage when pinning against an old server
A pin created in the new UI while the server is still pre-upgrade was lost
on the server upgrade. The pin toggle PATCHes `omnigent.pinned`; an old
server (no per-user pin concept) stores it as a bare label, but the
upgraded server's read path (`_labels_for_viewer`) drops every bare/
`omnigent.pinned.*` key and only surfaces the caller's own
`omnigent.pinned.<user>` key — so the bare-key pin silently vanishes. The
localStorage→server migration couldn't recover it either, since that pin
was never in localStorage.
This complements the earlier migration-gate fix (which protected pins made
*before* the UI upgrade). Now the toggle also checks `filterHonored`: when
the server can't store pins, it writes the pin to localStorage (the same
store the pre-upgrade UI used) instead of PATCHing a doomed bare key. The
pin renders immediately (sidebar unions localStorage pins) and later
migrates through `useMigrateLocalPinsToServer` like any pre-upgrade pin.
Once the server can store pins, the toggle uses the server as before.
- Move the legacy-pin localStorage helpers from Sidebar.tsx to the leaf
sidebarNav module (+ a single-id `setLegacyPinnedConversationId`) so the
toggle hook can use them without an import cycle.
- Tests: unit coverage for the toggle's old-server fallback (pin/unpin to
localStorage, no PATCH; normal PATCH path once honored), and an
end-to-end case in the backwards-compat suite that pins DURING the
UI-before-server window and asserts it survives the server upgrade.
Co-authored-by: Isaac
* fix(sessions): surface local-write failures in the old-server pin fallback
Addresses a review note: the old-server pin toggle's localStorage write is
the pin's only persistence, but it went through the best-effort
`writeLegacyPinnedConversationIds`, which swallows write errors (e.g.
storage quota exceeded). So a failed write let the mutation report success
and the optimistic patch show the pin, while it silently vanished on reload
— with no rollback.
Split out a throwing `...OrThrow` raw write. The old-server fallback
(`setLegacyPinnedConversationId`) now uses it, so a failed write rejects the
mutation → `onError` rolls back the optimistic patch and the UI honestly
shows the pin didn't take, matching the server PATCH path. The migration's
best-effort write is unchanged (a failed write there just retries next load).
Test: the fallback rolls back the optimistic pin when the local write throws.
Co-authored-by: Isaac
## Related issue
N/A
## Summary
- Add `omnigent/onboarding/sandboxes/types.py` with shared dataclasses
(`SandboxCapabilities`, `SandboxSpec`, `SandboxInfo`, `HostContext`) and the
new `SandboxError` exception hierarchy.
- Add `omnigent/onboarding/sandboxes/registry.py` with a contribution-based
provider registry that mirrors `omnigent/harness_plugins.py`: built-in
providers are declared as a `SandboxProviderContribution`, community
packages register via the `omnigent.sandbox_providers` entrypoint group, and
broken plugins are recorded in `load_errors` without breaking core startup.
- Add `omnigent/community/sandbox/__init__.py` as a namespace package so
third-party providers can ship code under `omnigent.community.sandbox.*`.
- Validation enforces that community provider code lives under the community
namespace, rejects name collisions, and checks metadata consistency.
- Add a `capabilities` property to `SandboxLauncher` that derives feature flags
from existing class variables and overridden transport methods.
- Migrate CLI and managed-host call sites from direct class-var reads
(`supports_cli_bootstrap`, `can_resume`, `supports_local_port_forward`) to
the new `capabilities` object.
- Add unit tests for types, registry behavior, validation, and entrypoint
discovery.
No provider implementations were changed; this is purely a surface-layer
refactor toward a pluggable sandbox provider interface.
## Test Plan
```bash
uv run pytest tests/onboarding/sandboxes tests/sandbox tests/cli/test_cli.py tests/server -k managed -q
pre-commit run --files omnigent/onboarding/sandboxes/types.py omnigent/onboarding/sandboxes/registry.py omnigent/onboarding/sandboxes/base.py omnigent/onboarding/sandboxes/__init__.py omnigent/onboarding/sandboxes/bootstrap.py omnigent/community/sandbox/__init__.py omnigent/cli_sandbox.py omnigent/server/managed_hosts.py tests/onboarding/sandboxes/test_types.py tests/onboarding/sandboxes/test_registry.py
```
All 779 selected tests pass and the targeted pre-commit hooks pass.
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [ ] Manual verification completed
- [x] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
New unit tests in `tests/onboarding/sandboxes/test_types.py` and
`tests/onboarding/sandboxes/test_registry.py` exercise the registry,
contribution validation, types, and capabilities derivation. Existing
provider and CLI tests pass unchanged, confirming backward compatibility.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
* feat(scheduled): add Model + Reasoning-effort pickers to the task dialog
The scheduled-task create/edit dialog previously omitted model and effort,
sending only agent_id so tasks always ran with the agent's configured
defaults. Add lightweight Model + Reasoning-effort controls, gated by the
selected agent's capability exactly like the interactive New Chat dialog:
they render only for native coding agents that carry the model/effort
surface (Claude Code) and are hidden for agents without it (Codex, plain
SDK agents, etc.).
- New scheduled-local ModelEffortFields component reuses the shared option
lists (CLAUDE_NATIVE_MODELS + the version-agnostic aliases, and
CLAUDE_NATIVE_EFFORTS) rather than importing the 26-prop
HarnessConfigModal, which is bound to smart-routing / cost-control /
per-turn model loading and disproportionate for a saved task. When a host
is pinned it uses that host's live model options; with none pinned (the
common case) it falls back to the static Claude aliases.
- Hoist CLAUDE_NATIVE_EFFORTS into the shared HarnessConfigControls module
so both dialogs share one source of truth.
- Wire modelOverride + reasoningEffort through create and update (both
already round-tripped by scheduledTasksApi.ts — no client/API change).
Unselected ("Default") omits the field on create so the fire path uses
the agent's defaults; on edit, Default sends null to clear a prior
override. Edit mode prefills both controls from the loaded task.
No permission/approval/cursor mode picker and no new API field: this is a
pure frontend change.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* test(automations): e2e for model + effort selectors
Extend tests/e2e_ui/scheduled/test_scheduled_tasks_page.py with UI journeys
for the model + reasoning-effort selectors added to the scheduled-task
create/edit dialog:
- controls visible + default to "Default" for a capability-gated agent
(Claude Code)
- controls hidden (with the "uses defaults" hint) for a non-capable agent
(seeded Codex task, asserted via the edit dialog)
- create persists a concrete Model + Effort pick (asserted via the REST API)
- create with both controls left on Default persists null overrides
- edit prefills the controls from a seeded task's stored overrides
LLM-free like the sibling tests: exercises only the dialog, REST, and the
rendered row. Uses Playwright expect() auto-waiting, no sleeps.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
---------
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
## Related issue
N/A
## Summary
- Remove the dedicated Settings → Appearance → Sidebar → Font size card and the `lib/sidebarFontPreferences` module, since users should use the global Interface font size control instead.
- Clear the legacy `omnigent:sidebar-font-size` localStorage key on app boot so anyone who previously changed the sidebar font size falls back to the default 13px.
- Add a "Reset to defaults" button at the bottom of the Appearance section that opens a confirmation dialog and resets all appearance choices: mode, terminal theme, color palette/custom theme, workspace panel default, hide-unconfigured-harnesses toggle, and interface/code font size and family.
## Test Plan
- Updated unit tests in `web/src/pages/SettingsPage.test.tsx` covering the reset flow and the absence of the sidebar font size control.
- Added a Playwright E2E test in `tests/e2e_ui/sessions/test_appearance_reset.py` to verify the sidebar card is gone and the reset dialog restores defaults.
- To verify locally after installing web dependencies:
- `cd web && npm run type-check`
- `npx vitest run src/pages/SettingsPage.test.tsx`
- `pytest tests/e2e_ui/sessions/test_appearance_reset.py`
## Demo
N/A — UI change; a screen recording of the reset confirmation dialog is recommended before merge.
## Type of change
- [ ] Bug fix
- [ ] Feature
- [x] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [x] E2E tests added / updated
- [ ] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
local web dependencies are not installed in this environment, so the local type-check and vitest runs could not be executed. CI will run the web test suite on the PR branch.
## Changelog
Settings → Appearance no longer has a separate Sidebar font size control, and a new "Reset to defaults" button restores every appearance preference after confirmation.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
* feat(projects): name the project in the new-session hero, drop the tray chip
When starting a session from within a project (a `?project=` landing), the
composer used to show the project as a pill in the footer tray while the hero
kept its generic "What should we do?" prompt. Move that context into the hero
instead: the heading shows the project name and Otto's eyes are swapped for the
same folder icon the sidebar uses for a project. The footer project chip
(`LandingProjectPicker`) is removed — filing on create still uses the same
`selectedProject` state, just without the redundant chip.
The folder icon renders in a fixed-height (`h-18`) box matching Otto so the
vertically-centered composer doesn't shift when toggling between the plain and
in-project landings.
Co-authored-by: Isaac
* fix(projects): clamp long project name in the new-session hero
A 100-char project name (the server-side cap) rendered at text-3xl overflowed
the centered container: the icon+heading flex row sized to its content with no
width bound, so the h1's min-w-0/line-clamp had nothing to act against. Give the
row w-full and keep the heading min-w-0 + line-clamp-2 + break-words so a long
name wraps to two lines and ellipsizes instead of overflowing. Add a test
asserting the clamp class contract on a 100-char name.
Co-authored-by: Isaac
The same-second filename collision was disambiguated by pid alone. A pid is
only unique across processes — a process that crashed more than twice within
one second reused its own pid, so every report after the first collision was
written to the same path and silently destroyed its predecessor. Saving five
reports in one second left two files on disk with three crash reports lost,
with rotation held wide enough that nothing should have been pruned.
Keep counting past the pid-suffixed name until the path is free.
test_save_report_writes_and_rotates encoded the bug: it asserted all five
returned paths still existed while rotation kept only two, which could only
hold when the collision collapsed them onto two names. It now asserts the
newest report survives its own rotation pass, and a new test pins the
no-overwrite guarantee with rotation held wide.
Signed-off-by: apeltekci <andrew@peltekci.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Swap the Files right-rail tab glyph from FilePenLineIcon (pen-on-page) to
FilesIcon (stacked pages) to better convey the panel's contents.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(sessions): don't wipe local pins when UI upgrades before server
The one-time localStorage→server pin migration (#3189) trusted an
ambiguous success signal. A pre-upgrade server silently ignores the
unknown `?pinned=true` param and returns the normal (unfiltered) session
page, so the UI saw ~100 "server pins", computed an empty to-migrate set,
and cleared localStorage without ever writing a pin. After the server was
upgraded, its per-user key filter found nothing and every pin read as
unpinned — the reported data loss for UI-before-server upgrades.
Fix, entirely client-side:
- `fetchPinnedConversations` now returns `{ conversations, filterHonored }`.
It keeps only rows actually carrying the `omnigent.pinned` label and
reports `filterHonored: false` when the server returned unpinned rows —
the tell-tale of an old server that ignored the filter.
- The migration is gated on `filterHonored`: it stays inert (localStorage
untouched) against an old server and re-runs after the eventual upgrade.
A legacy id is dropped only after its write is confirmed.
- Pinned membership is the union of the server's pins and any leftover
localStorage pins, so a not-yet-migrated pin keeps rendering instead of
vanishing during the UI-before-server window.
Tests: new filter-honored detection cases, a migration-gate suite, and an
end-to-end backwards-compat test that drives the real hooks across an
old→new server upgrade and asserts the pin is never lost.
Co-authored-by: Isaac
* docs(sessions): address Polly review notes on pin migration
- Document the empty-page ambiguity in `filterHonored` and why it's safe
(an old empty page means a zero-session account; the migration PATCH to a
deleted session 404s and the pin is retained, not lost).
- Note the window-scoped caveat that a legacy-only pin outside the loaded
paginated window may not render a row until loaded.
- Add a regression test: a failed (404) migration write keeps the legacy
pin in localStorage for retry.
Co-authored-by: Isaac
* feat(automations): absolute next-run time + card rows
Change 1: the Automations list now shows the next run as an absolute
wall-clock time ("Next run Tomorrow at 8:00 AM" / "Today at 2:30 PM" /
"Jul 26, 8:00 AM") instead of a relative delta ("in 15h"). Adds
formatNextRunAtAbsolute() in scheduleText.ts, which only FORMATS the
server-authoritative next_run_at (rendered in the task timezone,
Today/Tomorrow bucketed in that same zone) and never recomputes which
instant is next on the client. The old relative formatNextRunAt() is
kept intact.
Change 2: each ScheduledTaskRow now renders as a card (rounded-xl
border bg-card, internal padding), and TasksPage stacks them with a
gap. All existing behavior and data-testids preserved; paused rows are
not dimmed.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* feat(automations): relative full-word next-run label
Reverses the earlier absolute-time next-run display back to a
server-sourced relative delta in full words ("Next run in 3 hours",
"Next run in 8 mins", "Next run in 2 days") per user feedback.
formatNextRunAt now emits full-word, pluralized buckets ('soon' /
'in N min(s)' / 'in N hour(s)' / 'in N day(s)'); the delta is still
computed only from the server's authoritative next_run_at, so the
"no client countdown" rule is unaffected. Removes the now-dead
formatNextRunAtAbsolute and its private helpers (safeFormat,
civilDayInZone). Card-row styling is unchanged.
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* fix(automations): live-tick the relative next-run label
The relative next-run label was frozen at its first-render `now` and
only refreshed on remount. A shared 30s useNow() clock (a module-level
singleton via useSyncExternalStore) now drives live re-renders, so the
delta counts down while the page stays open. TasksPage owns the one
ticker and passes `now` to each row, keeping the row a pure function of
props.
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* fix(automations): round next-run label to nearest unit
Flooring understated the relative next-run label near a unit boundary:
a task 1h49m away read "in 1 hour". formatNextRunAt now rounds to the
nearest minute/hour/day and promotes on carry (each threshold tests the
already-rounded value), so 1h49m reads "in 2 hours" and a delta that
rounds up to a full unit shows "in 1 hour"/"in 1 day" rather than
"in 60 mins"/"in 24 hours".
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* test(automations): e2e for live-ticking next-run countdown
Adds a Playwright test to the scheduled-tasks page suite proving the
relative next-run label re-renders on its own as time passes (the shared
useNow() ticker), with no navigation. Uses clock mocking for determinism:
pins the browser clock 40 min before the server's next_run_at, asserts
"Next run in 40 mins", fast-forwards 35 min past many 30s ticks, then
asserts the same row updated to "Next run in 5 mins". LLM-free.
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
---------
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
## Related issue
N/A
## Summary
- When `omnigent server` binds a non-loopback interface, it auto-enables accounts (login) mode and prints a warning.
- The warning now explicitly names `OMNIGENT_AUTH_ENABLED=0` as the override to keep single-user mode.
- Kept the warning to the canonical env var; removed any mention of the deprecated alias.
- Improved the rendered indentation so the override sentence starts on its own line.
## Test Plan
- `uv run ruff check omnigent/cli.py`
- `uv run pytest tests/cli/test_bind_auth_defaults.py -q`
Both pass.
## Demo
N/A
## Type of change
- [ ] Bug fix
- [x] Feature
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [ ] Manual verification completed
- [x] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
The existing `tests/cli/test_bind_auth_defaults.py` already exercises the non-loopback auto-enable path and the explicit `OMNIGENT_AUTH_ENABLED=0` override. This change only updates the warning copy.
## Changelog
`omnigent server` now tells users they can set `OMNIGENT_AUTH_ENABLED=0` to override automatic multi-user mode when binding to a non-loopback interface.
## Related issue
N/A
## Summary
- Remove the long-deprecated `OMNIGENT_ACCOUNTS_ENABLED` environment-variable alias for the multi-user auth enable switch. The canonical name `OMNIGENT_AUTH_ENABLED` has existed since the repository was open-sourced.
- Strip the alias logic from `omnigent/server/auth.py::_auth_enabled()`, the explicit-auth check in `omnigent/cli.py::_apply_bind_auth_defaults()`, and the runner env-propagation allowlist in `omnigent/host/connect.py`.
- Delete the tests that exercised the alias and the obsolete comment in `tests/conftest.py`.
## Test Plan
- `uv run ruff check omnigent/server/auth.py omnigent/cli.py omnigent/host/connect.py tests/conftest.py tests/cli/test_bind_auth_defaults.py tests/server/test_accounts.py tests/e2e/test_local_server_lifecycle_e2e.py` passed.
- `uv run pytest tests/cli/test_bind_auth_defaults.py tests/server/test_accounts.py -q --no-header` passed (95 items).
- `uv run pytest tests/server/test_accounts.py::test_resolve_auth_source_defaults_to_header tests/server/test_accounts.py::test_resolve_auth_source_opt_in_selects_accounts tests/server/test_accounts.py::test_factory_defaults_to_header_when_env_unset tests/cli/test_bind_auth_defaults.py -q --no-header` passed (15 items).
- Verified no remaining references with `grep -R "OMNIGENT_ACCOUNTS_ENABLED" . --exclude-dir=.git --exclude-dir=.venv`.
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [x] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [x] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Removed the tests that specifically covered the deprecated alias; remaining tests continue to validate `OMNIGENT_AUTH_ENABLED` behavior. The refactor does not change the `OMNIGENT_AUTH_ENABLED=1 | =0` semantics.
## Changelog
[Breaking] The deprecated `OMNIGENT_ACCOUNTS_ENABLED` environment variable has been removed; use `OMNIGENT_AUTH_ENABLED` instead.
BREAKING CHANGE: Users and deploys still setting `OMNIGENT_ACCOUNTS_ENABLED` must rename the variable to `OMNIGENT_AUTH_ENABLED` before upgrading; the old name is no longer read or propagated.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
Bring the Projects PRD implementation-status section in line with what has
shipped and what remains:
- Mark backend `config` hardening (size bound + non-dict coercion) as done —
both already landed in the project store.
- Move the completed Benchmark (#3094) and Phase 2 (project defaults) items out
of TODO into their own "Done" sections.
- Correct a stale claim that the new-session prefill machine still reads the
`omni_project` label — it was collapsed to config-only in Phase 2. The one
remaining UI label reader (the Settings archived-project picker) is folded
into the Phase 4 retire-label-path step instead.
- Postpone Phase 3 (memory & context) and Phase 4 (label consolidation) with
distinct triggers: Phase 3 waits for customer demand; Phase 4 waits until
telemetry shows most clients have migrated to a version that writes
`project_id`.
Co-authored-by: Isaac
## Related issue
N/A
## Summary
- Updated `inject_user_message()` in `omnigent/claude_native_bridge.py` so user messages that start with a Claude Code UI-only/unsupported slash command (`/help`, `/exit`, `/quit`, `/doctor`, `/cost`, etc.) are escaped before being pasted into the TUI.
- Escaping inserts an invisible zero-width no-break space before the leading `/`, causing Claude Code to treat the input as regular user text while the user still sees their slash.
- Supported slash commands (`/clear`, `/compact`, `/effort`, `/model`, `/ultrareview`, `/branch`, `/fork`) and unknown skill commands pass through unchanged.
## Test Plan
- Added parametrized unit test for `_escape_unsupported_slash_command`.
- Added payload test verifying `/help` gets the escape prefix and `/clear` does not.
- Ran targeted injection tests and pre-commit:
- `uv run pytest tests/test_claude_native_bridge.py::test_escape_unsupported_slash_command tests/test_claude_native_bridge.py::test_inject_user_message_escapes_unsupported_slash_command_payload -q`
- `uv run pytest tests/test_claude_native_bridge.py -k "inject_user_message" -q`
- `uv run ruff check omnigent/claude_native_bridge.py tests/test_claude_native_bridge.py`
- `uv run ruff format omnigent/claude_native_bridge.py tests/test_claude_native_bridge.py --check`
- `uv run pre-commit run --files omnigent/claude_native_bridge.py tests/test_claude_native_bridge.py`
## Demo
N/A
## Type of change
- [x] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [ ] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
N/A — new unit tests directly cover the escaping decision and the payload path.
## Changelog
Unsupported Claude Code slash commands are now escaped and sent as regular user messages instead of leaving the native terminal in an undriveable state.
* fix(projects): disable settings inputs during config load; correct worktree doc
Two non-blocking follow-ups from the PR #3221 review:
- Gate the worktree toggle, workspace Browse trigger, and path input on
`isLoading`, matching the host Select. Previously an edit made in the load
window would be clobbered by the seeding effect once the fetch settled.
- Rewrite the `use_worktree` docstring to match the opt-in implementation
(only `true` is written; `false` is never stored and treated as unset).
Co-authored-by: Isaac
* docs(projects): mark backend config hardening as done in PRD
The two #3108 config-hardening follow-ups (size bound + non-dict coercion)
already landed in the project store; move them from "deferred" to a ✅ bullet
so the PRD status matches the code.
Co-authored-by: Isaac
* feat(projects): project settings editor + config-driven composer prefill (Phase 2)
Add a "Project settings" dialog to set a project's stored session defaults
(host, working directory, agent, opt-in random worktree) and wire the new-chat
composer to prefill from that stored config, retiring the newest-session
inference so stored config is the single source of truth.
- ProjectSettingsDialog: edit + persist config {host_id, workspace, agent_id,
use_worktree}; worktrees opt-in (default OFF, store true when on). Reuses the
composer's host/agent pickers and filesystem browser.
- projectPrefill: collapse to config-only seeding; unset fields fall through to
the composer's generic defaults. Honor a stored sandbox default via
selectSandbox (gated on managed sandboxes). Remove useNewestProjectSession.
- Extract the nested-dropdown dismiss guard into a dependency-free module shared
by the settings and scheduled-task dialogs.
Co-authored-by: Isaac
* fix(projects): repair CI — Sidebar test mocks, e2e rewrites, retire inference e2e
- Add useProjectConfig/useUpdateProjectConfig to all 10 Sidebar test mocks
(Sidebar now mounts ProjectSettingsDialog, which calls them).
- Rewrite the settings-dialog e2e to create the project via POST /v1/projects
instead of the flaky row-kebab move-to-project flow.
- Fix the composer-prefill e2e to stub GET /v1/sessions/projects (bare array),
the real endpoint useProjects hits.
- Remove test_start_session_project_prefill — it exercised the newest-session
inference path this PR retired; config-driven prefill replaces its coverage.
Co-authored-by: Isaac
* fix(projects): address review — no data-loss on failed config load; fresh prefill after save
Blocking issues from the PR review:
1. Data loss: saving the settings dialog after a failed config GET sent `{}`,
which the server reads as "clear stored defaults". Now `useProjectConfig`'s
isError is surfaced; a first-class project whose config failed to load blocks
Save (with a notice), the seed effect skips a blank draft, and onSubmit bails.
2. Stale prefill after save: useUpdateProjectConfig only invalidated, so the
composer's one-shot prefill could latch onto a stale cached config (30s
staleTime) and drop just-saved defaults. It now setQueryData's the fresh
config and upserts the projects list (so a promoted label-only folder
resolves to its new id immediately).
Tests: dialog load-error blocks Save; hook seeds config + upserts list on
success; useProjectConfig disabled on null id and surfaces isError.
Co-authored-by: Isaac
* fix(web): center sidebar header buttons and soften session row hover
## Related issue
N/A
## Summary
- Vertically center section header action buttons (Projects `+`, Sessions kebab, etc.) with their titles by using `top-1/2 -translate-y-1/2` instead of `top-0.5`.
- Remove the 1 px lift on session row hover (`motion-safe:hover:-translate-y-px`) so rows stay visually anchored.
- Calm the hover flash by dropping the bouncy Otto-token transition on rows and reducing the global `--sidebar-hover` tint from 5% to 3%. Rows now use the same plain `transition-colors` pattern as the rest of the sidebar hover surfaces.
- Make `SIDEBAR_ACTIVE_HIGHLIGHT` also specify `:hover` styles so active items (current page, selected session, drop target) keep their active background on hover instead of switching to the hover tint.
## Test Plan
- `cd web && npm install && npm run dev`
- Hover over Projects/Sessions headers and confirm action buttons are vertically centered with the title text.
- Hover over active items (e.g., current page in the top nav, selected session row, current Inbox) and confirm the background stays in the active state and does not flash.
- Hover over inactive session rows and confirm the row no longer shifts up and the background highlight is subtler.
## Demo
Subtle hover/positioning polish. Verify by hovering items in the sidebar — buttons align with title baselines, rows stay still on hover, and active items don't flash.
## Type of change
- [x] Bug fix
- [x] UI / frontend change
- [ ] Feature
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Visually verified by inspecting the relevant Tailwind classes and CSS variables. No test coverage changes; the existing `Sidebar.projectHeaderChevron.test.tsx` covers header layout, and the hover behavior is primarily CSS.
## Changelog
Sidebar header action buttons are now vertically centered with section titles. Session row hover is smoother, and active items no longer flash when hovered.
* test(e2e-ui): regenerate visual baselines
---------
Co-authored-by: omnigent-ci[bot] <294685417+omnigent-ci[bot]@users.noreply.github.com>
## Related issue
N/A
## Summary
- Replace the web-resolved-theme bridge with a single cross-shell contract, `setThemeSource(theme)`, so the web app only reports the user's chosen theme source and each shell drives its own OS-level dark mode.
- Android: `MainActivity` now extends `AppCompatActivity`; `OmnigentBridgeListener` maps `setColorScheme` to `AppCompatDelegate.setDefaultNightMode`; system-bar icon contrast is derived from `resources.configuration.uiMode`. Removes `ResolvedColorScheme.kt`, the root-class MutationObserver, and the top-level navigation reset on init.
- iOS: Add a `ThemeSource` enum and `ThemeController` singleton inside the existing `OmnigentWebView.swift` target file to avoid `.pbxproj` edits; wire `setColorScheme` through the JS bridge and apply it via `.preferredColorScheme(...)` and `window.overrideUserInterfaceStyle`.
- Web: Update `nativeBridge.setThemeSource`, remove the `omnigent-native-ready` queue, and update `ThemeProvider`/`nativeBridge` unit tests.
- Android and web unit tests are updated to match the new contract.
## Test Plan
- iOS: `cd web/ios && xcodebuild -project Omnigent.xcodeproj -scheme Omnigent -destination 'platform=iOS Simulator,name=iPhone 17 Pro' -configuration Debug -only-testing:OmnigentTests test`
- Android: `cd web/android && ./gradlew :app:testDebugUnitTest`
- Web: `cd web && npm install && npm run type-check && npm run test -- ThemeProvider.test.tsx nativeBridge.test.ts`
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Manual verification completed for iOS: the app builds and `OmnigentTests` passes in the iPhone 17 Pro simulator. Android and web test suites were not run end-to-end in this session, but the affected unit tests were updated in the same change.
## Related issue
N/A
## Summary
- Add a top-level `justfile` that groups common local dev tasks (`run-ios`, `run-android`, `dev`, `electron-dev`, `lint`, `normalize-locks`, etc.) with hidden `_ensure-*` / `_check-*` prerequisites.
- Add an iOS `simulator` Fastlane lane that builds the Debug .app, installs it on an already-created iOS Simulator, and launches it.
- Add Android Gradle tasks (`runDebug`, `reverseProxy`) for launching the debug APK and running `adb reverse`.
- Fix the Fastlane `xcodebuild` invocation to use camel-case `derivedDataPath` so the built `.app` is written where the lane expects it.
- Export `FASTLANE_SKIP_UPDATE_CHECK=1` in the justfile.
- Document the new `justfile` recipes concisely in `AGENTS.md`.
## Test Plan
- `just --list` shows grouped recipes.
- `just run-ios` built/launched the iOS app in the iPhone 17 Pro Simulator.
- `pre-commit` passes on the touched files.
## Demo
N/A
## Type of change
- [x] Feature
- [ ] Bug fix
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [ ] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Verified manually by running `just run-ios` and watching the Omnigent app launch in the iOS Simulator.
## Changelog
Add a top-level `justfile` with recipes for launching the iOS Simulator, running the Android debug build, starting the omnigent dev pod, and running pre-commit/lockfile normalization.
* feat(web): 3D model preview for STL / 3MF / OBJ files
Selecting an .stl / .3mf / .obj file in the Files browser now renders an
interactive WebGL preview (orbit/zoom/pan) instead of the "Preview not
available for binary files" placeholder.
- Add `isModelFile()` to codeViewerHelpers (MIME-first, extension fallback),
scoped to exactly STL/3MF/OBJ.
- New lazy-loaded `ModelViewer` component (three.js STLLoader/3MFLoader/
OBJLoader) with camera + OrbitControls, lighting, auto-fit, loading/error
states, and full scene teardown on unmount.
- Dispatch models before the binary-rejection branch in CodeViewer; treat
them like images in FileViewer (diff/source-mode suppressed).
- three.js pinned at 0.185.1 and code-split into its own chunk so it stays
out of the main bundle.
Co-authored-by: Isaac
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(web): address model-viewer review — unified resolver, recovery, teardown
Resolve the four blocking issues from cross-vendor review of the 3D model
preview:
1. Unified format interface: add one shared `getModelFormat(path, contentType)`
resolver (MIME-first, extension fallback) used by BOTH `isModelFile`
dispatch and `ModelViewer`'s loader selection, so a MIME-matched file with
an unknown extension parses via the correct loader instead of erroring.
`isModelFile` is now `getModelFormat(...) !== null`.
2. Error state no longer unmounts the canvas: the container is always mounted
and the error is an overlay on top, keeping the ref alive so an
invalid→valid prop change recovers.
3. Single idempotent `teardownScene()` called from both the init failure path
and the effect cleanup, so a partial init (renderer/controls/context/RAF)
can't leak on failure.
4. Empty/degenerate models (e.g. comment-only OBJ) are validated for a
non-empty, finite bounding box before fitting; invalid bounds route to the
error UI instead of a blank canvas.
Adds ModelViewer.test.tsx (MIME-only loader selection, malformed/empty/NaN →
error, invalid→valid recovery, failure-path + unmount teardown) and
getModelFormat unit tests.
Co-authored-by: Isaac
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* feat(web): theme-aware 3D model preview (light/dark)
ModelViewer previously hardcoded a neutral STL material, fixed light
intensities, and a transparent canvas, so the 3D preview ignored the app
theme. Make it theme-aware off the SAME next-themes source Monaco and the
terminal use (`useTheme().resolvedTheme`), so it tracks light/dark and
updates live when the user toggles the theme with a model open.
- Add a pure `modelViewerTheme(resolved)` map in codeViewerHelpers (mirrors
`resolvedThemeToMonaco`): background clear color, STL default material, and
ambient/key light intensities per mode — brighter lights in dark so the
mesh stays legible. Shared across STL/3MF/OBJ in the one unified pipeline.
- ModelViewer seeds the scene from the active mode and keeps light/material
handles on its resource bag so a theme toggle recolors the live scene in
place (clear color + intensities + STL color) with no reload/reparse.
- Drop the transparent (alpha) canvas in favor of a theme-derived opaque
background so the preview sits flush with the panel in both themes.
- Tests: three theme-awareness cases (light build, dark build, live toggle
without rebuild) mirroring the next-themes mock pattern in
MonacoCodeEditor.test.tsx, plus modelViewerTheme unit tests.
Co-authored-by: Isaac
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* test(web): 3MF MIME-only dispatch + prune package-lock churn
Add MIME-only 3MF coverage mirroring the existing STL/OBJ tests: a file
with an absent/unrecognized extension but a `model/3mf` content type must
resolve to the 3MF loader in ModelViewer and route to <ModelViewer> in
CodeViewer, exercising the shared getModelFormat() resolver.
Regenerate web/package-lock.json so the diff vs origin/main is limited to
the `three` dependency subtree — dropping unrelated resolved-URL
normalization churn from an earlier regen.
Co-authored-by: Isaac
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(web): dispose material textures in ModelViewer teardown
disposeObject() freed each mesh's geometry and material but not the
textures the material references (map, normalMap, roughnessMap, …), so a
textured 3MF leaked its GPU textures every time the viewer unmounted.
three.js frees neither the material nor its textures automatically.
Add disposeMaterial(), which disposes every texture slot on a material
(detected via the three.js `isTexture` flag, robust to multiple three
copies) before disposing the material itself. Extend the ModelViewer
teardown unit test with a textured-material mesh and assert its textures
are released on unmount.
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* test(e2e): cover 3D model preview in the Files browser
Add a Playwright e2e test that seeds an ASCII STL and an OBJ file, opens
each in the Files browser, and asserts the ModelViewer mounts: the
`3D preview of …` canvas host renders a <canvas>, the "Unable to render
3D model" overlay never shows (so parsing and WebGL both succeeded), and
the flow does NOT fall through to the binary placeholder or a source
view. STL exercises MIME-based routing (application/vnd.ms-pki.stl); OBJ
exercises the extension fallback. Seeded via the filesystem PUT endpoint
(no agent run), mirroring the existing image/pdf rendering e2e tests.
This satisfies the E2E UI Required gate for the model-preview feature.
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(web): resolve the 3D-viewer deps from the public npm registry
The three.js stack this PR added (three, @types/three,
@dimforge/rapier3d-compat, @tweenjs/tween.js, @types/stats.js,
@types/webxr, fflate, meshoptimizer) was locked with `resolved` URLs
pointing at an internal mirror (npm-proxy.dev.databricks.com), while the
rest of package-lock.json resolves from registry.npmjs.org. Public CI
can't reach that mirror, so `npm ci` timed out fetching
three-0.185.1.tgz (ETIMEDOUT) and failed the install-dependent checks.
Repoint just those eight `resolved` URLs to the canonical
registry.npmjs.org form. Integrity hashes are unchanged (the mirror
served identical tarballs), so this only changes where the tarballs are
fetched from, not what is installed. `npm ci --legacy-peer-deps` now
succeeds from a clean node_modules, and `npm install --package-lock-only
--legacy-peer-deps` produces no further diff, so the lockfile-up-to-date
gate stays green.
Signed-off-by: Bryan Li <bryan.li@gmail.com>
---------
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(android): honor system dark mode
Signed-off-by: Bryan Li <15131870+btli@users.noreply.github.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(android): sync system bar contrast
Signed-off-by: Bryan Li <15131870+btli@users.noreply.github.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(android): harden resolved theme sync
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* refactor(android): decode theme at bridge boundary
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* test(android): tighten theme bridge coverage
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(android): drop WebView algorithmic darkening
Algorithmic darkening inverts the SPA when the user forces light mode
while the OS is dark: the page's root color-scheme is then 'light', so
WebView treats it as dark-unaware and darkens it algorithmically,
leaving dark status-bar icons over a darkened page. With targetSdk >= 33
the DayNight host theme alone makes prefers-color-scheme track the OS,
so the darkening flag added nothing for the system-mode path and only
broke the forced-light path. Verified on an API 34 emulator across the
OS-light/dark x app-System/Light/Dark matrix.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(web): keep Electron on the selected theme, not the resolved one
Reporting only resolvedTheme regressed Electron system mode: an explicit
Light selection under a light OS changes no resolved value, so no report
fired and themeSource stayed 'system' — the shell chrome then flipped
dark with the OS while the app was forced light. Report the resolved
scheme first (Android system-bar contrast) and follow with 'system'
while that is the selection: Electron keeps the last report, so it
tracks the OS in system mode and pins to explicit selections, including
ones that leave resolvedTheme unchanged. Android drops 'system' at the
bridge, so its behavior is unchanged.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix: route native themes by consumer
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(android): harden system bar theme sync
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* test(android): clean up theme bridge state
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(android): install theme bridge at document start
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(android): resync system bars on live theme changes
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* style(android): format theme test
Signed-off-by: Bryan Li <bryan.li@gmail.com>
---------
Signed-off-by: Bryan Li <15131870+btli@users.noreply.github.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
Co-authored-by: Bryan Li <15131870+btli@users.noreply.github.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
github-release.yml fired on every v[0-9]* tag push and created an
unpublished DRAFT release for rc/dev/alpha/beta tags. Nothing downstream
depended on those drafts — draft-release-notes.yml already skips rc,
finalize-release.yml refuses rc, and the Docker/homebrew/changelog
workflows fire on the tag push / release:published directly. The drafts
just accumulated (and rehearsal rcs had to be gh-release-deleted during
cleanup).
Add a guard that skips the draft-release job for rcN/devN/preN tags
(trailing digit required so a substring like 'dev' in a mistyped tag can't
trip it). Drop the now-dead alpha/beta arms — this repo only cuts rc
pre-releases — and align the same rc/dev/pre pattern + comments across
the other release-adjacent workflows for consistency. Update release.yml's
Next-steps text and RELEASING.md accordingly.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
## Related issue
Closes #[F-CR-7]
## Summary
- After a user consents to an unknown server (deep link) or types a server URL, `WorkspaceURLExpander.expandIfNeeded` issued a HEAD probe via `URLSession.shared` with no redirect policy, so the consented host could 3xx-redirect the probe to a different origin — including a local-network service — breaking the consent alert's promise that the app only talks to the host the user approved.
- The probe now defaults to a dedicated `URLSession` backed by `SameOriginRedirectHandler`, a `URLSessionTaskDelegate` that follows only same-origin redirects (scheme + host + port match) and blocks any cross-origin redirect by returning `nil` from `willPerformHTTPRedirection`.
- As defense in depth, `expandIfNeeded` additionally verifies `response.url`'s origin matches the approved origin, so a cross-origin response is never trusted even if a caller supplies a bare session without the redirect delegate.
- Rebased onto #3179 (F-CR-6) and deduped: removed my `--omnigent-deep-link` test hook (subsumed by #3179's `--omnigent-open-url` / `--omnigent-reset-state` seam), and consolidated the two `MockHTTPServer` copies into one shared file compiled into both test targets.
## Test Plan
- Unit: `WorkspaceURLExpanderTests.testRejectsResponseFromDifferentOrigin` returns a `server: databricks` 200 whose `url` is a different origin and asserts the URL is left unchanged.
- Integration (simulator, real local HTTP network): `WorkspaceURLExpanderRedirectTests.testBlocksCrossOriginRedirect` / `testFollowsSameOriginRedirect` assert a cross-origin redirect is blocked (response stays 302 on the approved port) and a same-origin redirect is followed. Confirmed meaningful: the cross-origin test fails when the delegate is reverted to follow-all-redirects (the vulnerable behavior).
- UI (simulator): `RedirectConsentUITests.testDeepLinkConsentOpensApprovedServer` drives the deep-link consent flow via #3179's `--omnigent-open-url` + `--omnigent-reset-state` seam and asserts the alert appears, "Open" loads the approved server's WebView.
- Ran on iPhone 17 simulator: all 8 expander/redirect tests + the UI smoke test + all 26 F-CR-6 deep-link tests pass; full project builds.
- Note: the UI test cannot exercise the redirect itself — a localhost deep link infers `http`, and the probe is https-only, so the probe never fires for loopback. The redirect policy is verified over a real local network by the integration test instead.
## Demo
N/A
## Type of change
- [x] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [x] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [x] Integration tests added / updated
- [ ] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Manual verification: built and ran the new unit, integration, and UI tests on the iPhone 17 simulator; confirmed all pass and that the integration test fails against the vulnerable (follow-all-redirects) baseline, proving it is a meaningful regression test. Also ran all F-CR-6 tests after the dedup to confirm no regression from #3179's shared seam.
## Changelog
The iOS app no longer follows cross-origin redirects when probing a newly approved server for the Databricks workspace mount, so a consented host can't redirect the probe to a different origin.
* test: stabilize two known flakes (dictation close, agent-info popover)
Two load-timing flakes that recur across PRs:
- Pytest (server-rest) test_dictation.py::test_stream_closes_take_on_
abrupt_disconnect: on an abrupt disconnect the route offloads
handle.close() to a thread. During teardown the loop's thread-pool
executor may already be shutting down, so the offload raises and the
old contextlib.suppress swallowed it — the take (and, for the remote
engine, a worker slot) leaks. Fall back to a direct close() on the
loop; it's a quick non-blocking free for every engine.
- E2E UI test_agent_info_popover.py: _open_popover single-clicked the
trigger, but the button hover-opens on the click's own pointer arrival
and the click's Radix toggle can flip it back shut past the
HOVER_CLICK_GRACE_MS window under load, so the panel never mounts.
Confirm the panel opened and retry the click from a closed state.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* test: stabilize scheduled-tasks time-picker flake
test_scheduled_task_create_edit_modal_and_time_picker had two coupled
races in the time-picker step (6/10 failures reproduced, no artificial
load needed):
- The picker is a Radix popover nested in the create-task dialog. The
dialog's focus management can fire an interaction-outside that closes
it the instant it mounts, so the minute cells unmount between the
visibility check and the click (element-not-found / click timeout).
- Selecting a minute leaves the popover open, and an open floating-ui
popover keeps recomputing its position — so the submit button (and,
later, the edit-phase time input) stays perpetually "not stable" and
detaches mid-click.
Extract a _pick_minute() helper that opens from a known-closed state and
retries until the cell is present, then dismisses the picker via a
click-outside (not Escape, which would bubble to the Radix Dialog and
close it) and waits for it to unmount so the layout settles before
submit. 0/12 clean + 0/8 under load after the fix.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
The in-session "Configure" model dropdown offered a "Smart Routing" option on
native terminal sessions (Claude Code, Codex, Pi, …). It's meaningless there:
a native CLI bakes its model into the launch argv once and can't per-turn
route, so picking it did nothing useful.
Add isNativeTerminalSession() (mirrors the server's
_native_coding_agent_for_session: native by omnigent.wrapper label OR resolved
harness) and exclude such sessions from costRoutingEligible in ChatPage, so the
Smart Routing option no longer appears in their Model dropdown. Brain-harness
sessions (claude-sdk / codex / pi, and the polly orchestrator) keep it.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(scheduled tasks): add windowed latest-run-status store query
Add ScheduledTaskStore.list_latest_run_status_for_tasks(ids) -> {id: status},
a single row_number()-windowed query (scheduled_at DESC, id DESC — same order
as list_runs) returning each task's most-recent run status. Powers the Tasks
list completion badge in one query instead of N per-row /runs fetches, and is
correct under overlapping run-now runs (unlike a denormalized last_run_status
column). Tasks with no runs are absent from the map.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* feat(scheduled tasks): run-now endpoint + status/next-run serializer fields
Backend for three Tasks-list run controls:
- last_run_status: _to_response now carries the task's most-recent run status
(from the windowed store query), populated on list/get/patch. Force-fail of
stale orphans runs BEFORE the status read so a dead run reports failed, not a
stuck running.
- next_run_at: _to_response carries the live scheduler's authoritative next-fire
ISO timestamp (scheduler.next_run_at) on list/get/create/patch — server-
sourced, never client-recomputed (paused/unarmed → null).
- POST /v1/scheduled-tasks/{id}/run: an immediate manual fire that REUSES the
shared fire path via build_run_now (same _run_fire_for_task body, dispatch/
preflight seams, and in-flight overlap guard as the scheduler). Paused tasks
are runnable (manual override); fire-and-forget → 202 Accepted. 409 when a
fire is already in flight, 404 for a non-owned task, 503 when the scheduler
subsystem is not running. Wired via app.state.scheduled_task_run_now.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* feat(scheduled tasks): status pill, run-now menu, next-run on rows
Wire the three run controls into the Tasks list UI:
- last_run_status → a completion pill on each row (Failed/Skipped/Running/
Queued). Succeeded and never-run render NO pill (success is not noise);
Failed is destructive, Skipped muted — matching the Paused pill styling.
- next_run_at → "Next: <time>" on the schedule subline, formatted in the
task timezone via a new formatNextRunAt() that only FORMATS the server's
ISO value (never client-recomputes; paused/unarmed → nothing).
- Run now → a "⋯ menu" item + useRunScheduledTaskNow mutation (POST
/{id}/run) that invalidates the list + that task's runs so the pill
updates. Runnable for paused tasks; row busy-disables while in flight.
scheduledTasksApi gains lastRunStatus + nextRunAt (interface + wire map)
and runScheduledTaskNow(). Unit tests: pill per status, no-pill cases,
next-run formatting (tz + calendar-day boundary), run-now mutation wiring.
e2e: new run-controls journey (Run now → recorded run + pill flips);
existing schedule-line assertions relaxed to to_contain_text now that the
server next-run renders on the same line.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* style(scheduled tasks): bump task title size/weight on rows
Make the scheduled-task row title slightly larger and bolder: text-sm →
text-base and font-semibold → font-bold. Subline, pills, and spacing are
unchanged. Updates the one TasksPage sort-order test that located the title
by its .font-semibold class to .font-bold.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* Revert "style(scheduled tasks): bump task title size/weight on rows"
This reverts commit e0195ce3c4977abdaeba5316426029f808edeef6.
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* style(scheduled tasks): title 15px, metadata 13px on rows
Trim the row title to exactly 15px and the metadata subline to exactly 13px
using arbitrary-px classes (text-[15px] / text-[13px]) — the app root scales
rem ~1.125×, so the standard text-sm/text-xs would render 15.75/13.5px and
can't hit the exact target. Weights unchanged: title font-semibold (600),
subline no weight class (inherits 400). Pills, spacing, next-run text, and the
⋯ menu are untouched.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* style(scheduled tasks): lighten metadata subline on rows
Soften the row metadata subline one notch to a lighter gray via an opacity
step on the same theme token: text-muted-foreground → text-muted-foreground/80.
Theme-aware (works in light + dark), size unchanged (13px), and the next-run
<span> keeps inheriting the same color (no own color class). Title, pills,
spacing, and the ⋯ menu are untouched.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* style(scheduled tasks): tighten row spacing 2px, remove run-status pill
- Row vertical padding py-3 → py-[11px]: trims each side 1px so the gap
between adjacent rows drops from 24px to 22px (the list is flex-col with no
gap, so the row padding is the whole inter-row spacing).
- Remove the last-run status pill (Failed/Skipped/Running/Queued) entirely per
design: drop the render block, the RUN_STATUS_PILL map, the statusPill local,
and the now-unused ScheduledTaskRunStatus import. The Paused pill is kept
as-is. The lastRunStatus API/store field is left in place (harmless data;
only the visual is removed). Subline, next-run text, and the ⋯ menu unchanged.
Drops the per-status pill test cases in ScheduledTaskRow.test.tsx (that UI is
gone); keeps the paused-pill, next-run, and run-now tests.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* style(scheduled tasks): relative "Next run in Xh" on rows
Switch the row next-run display from an absolute label ("Next: Today 9:00 AM")
to a compact relative delta ("Next run in 15h" / "in 6d" / "soon"):
- formatNextRunAt now returns a delta (nextRunAt − now): <60m → "in Xm" (min
"in 1m"), <24h → "in Xh", else "in Xd", all floored; a delta below 1 min
(imminent / clock skew) → "soon"; null/unparseable iso → null. The `timezone`
param is dropped (a pure delta needs no zone) — call site + useMemo deps
updated. This only formats HOW FAR AWAY the server's authoritative next_run_at
is; it never recomputes WHICH instant is next on the client, so the old
"no client-recomputed countdown" rule still holds.
- Row prefix "Next: " → "Next run " so it reads "Next run in 15h".
Tests: rewrote the formatNextRunAt unit tests for the relative buckets +
boundaries + "soon" + null; updated the row test to the "Next run in …" prefix;
reconciled the e2e (the old count==0 "Next run" guard flips to positively
asserting the server-derived relative label — its real intent, no client
recompute, is unchanged).
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
* style(scheduled tasks): darken row hover background
Bump the full-row hover tint one notch: hover:bg-muted/50 → hover:bg-muted/70
(same theme-aware `muted` token, higher opacity). The color-mix stays
`var(--muted) N% transparent`, so in light the effective tint goes ~2.9% → 4.1%
black and in dark the alpha goes 0.5 → 0.7 — visibly stronger but still subtle.
Comment updated to match. Nothing else changes.
Co-authored-by: Isaac
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
---------
Signed-off-by: Rahul Ravindranathan <rahulnathan73@gmail.com>
Adds a 'databricks_cli' credential_proxy type so sandboxed tools can use
the Databricks CLI without the real OAuth/PAT token ever entering the
sandbox. The operator lists which ~/.databrickscfg profiles to proxy;
each is materialized into the sandbox as a placeholder-only .databrickscfg
(oa_cred_* token), and the L7 egress proxy swaps the placeholder for the
real token on the way out.
- Refreshing token provider (DatabricksProfileTokenProvider) re-mints
short-lived OAuth tokens via the databricks SDK for long sessions;
CredentialRewriteRule gains an optional secret_provider and the proxy
resolves secrets per-swap (offloaded via run_in_executor).
- Placeholder-only files are materialized into the sandbox scratch dir
and pointed at via DATABRICKS_CONFIG_FILE / DATABRICKS_CONFIG_PROFILE.
- Requires the 'databricks' extra and linux_bwrap (the Go CLI ignores
SSL_CERT_FILE on macOS, so darwin_seatbelt is rejected at parse time).
- Egress stays operator-listed: the workspace host must be named in
egress_rules, consistent with the other credential_proxy types.
Signed-off-by: mxatone <mxatone@gmail.com>
## Related issue
N/A
## Summary
- Bring `omnigent-slack` into the lockstep release cycle (now four packages, not three): its `[project].version` was stuck at `0.1.0` while the rest of the repo moved to `0.7.0.dev0`, so the extra pin and lockfile drifted.
- Pin `omnigent-slack==0.7.0.dev0` in the root `slack` optional-dependency extra, mirroring the existing `omnigent-client==` / `omnigent-ui-sdk==` sibling pins so a published `omnigent[slack]` always pairs with the matching `omnigent-slack` release.
- Teach `scripts/update_versions.py` (the engine behind `.github/workflows/bump-version.yml`) about the 4th package: rewrite the slack `[project].version` and the extra `==` pin on every bump, and scan `[project.optional-dependencies]` (not just `[project.dependencies]`) when verifying sibling pins. Regenerate `uv.lock`.
## Test Plan
- `uv run --no-project --python 3.12 --with packaging python scripts/update_versions.py check` → prints `0.7.0.dev0` (all four packages agree, all sibling `==` pins present).
- `uv lock` → "Updated omnigent-slack v0.1.0 -> v0.7.0.dev0".
- `uv run ... python -m pytest tests/scripts/test_update_versions.py` → 13 passed (updated the test fixture + assertions for the 4th package).
- `tests/test_version.py::test_version_matches_pyproject` still passes (root pyproject == `omnigent/version.py` at `0.7.0.dev0`).
## Demo
N/A
## Type of change
- [ ] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [x] Refactor / chore
- [ ] Docs
- [ ] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [ ] Integration tests added / updated
- [ ] E2E tests added / updated
- [ ] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Updated `tests/scripts/test_update_versions.py` to include `integrations/slack/pyproject.toml` in the `repo_copy` fixture and adjusted the lockstep assertions (5 changed files, 4 `9.9.9` occurrences in root pyproject, 1 in slack). Verified the full suite (13 tests) passes. Also ran `update_versions.py check` and `uv lock` manually to confirm lockstep + lockfile consistency.
* feat(onboarding): report the installed-but-unconfigured harness state
Add the credential axis to the picker-facing readiness map so Claude, Codex,
and Pi report the yellow "installed but no credential" state — the signal the
web setup dialog needs to know when to offer an "Add credential" action
(the M2 keystone of Setup-From-the-UI). Purely additive: the values
("needs-auth" / "binary-missing") already exist in HarnessAvailability, and
the launch gate (harness_is_configured) stays binary-only, so a not-yet-authed
harness is never blocked from launching.
- New _family_provider_configured(): whether an omnigent-managed provider
(API key / gateway) serves the harness's family, reading the same config
omni setup's overview does. Subscription-kind is excluded (that lives in the
CLI's own login, judged by harness_cli_logged_in). Local, side-effect free,
never raises (fails to "no credential").
- Claude: ready when a provider is configured OR the CLI subscription login is
present (was CLI-login only — an API-key-only user wrongly showed yellow).
Checks config first to avoid the status subprocess on the common path.
- Pi: gains the axis it lacked entirely (was binary-only → always green once
installed). No CLI login, so binary + provider: installed-but-no-provider is
now "needs-auth".
Codex already had this (unchanged, the template). Qwen/OpenCode env-auth
unchanged.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* feat(onboarding): write a harness provider credential from the UI
Second PR of Setup-From-the-UI (M3, security-sensitive). Adds the path that
turns a yellow "installed but not configured" harness green from the browser
for Claude / Codex / Pi, host-agnostic (local or remote), reusing the
credential-write logic omni setup already uses.
Design: the server is an authz'd pass-through. It validates ownership + the
UI-auth allowlist and forwards the secret over the (TLS) tunnel; the host DAEMON
does the write on the runner. The server never persists the secret, and the
frame's secret_value field is redaction-named so it never lands on a telemetry
span. Gated behind OMNIGENT_HARNESS_INSTALL_ENABLED (default off) exactly like
the install route (404 when disabled).
- New non-interactive core omnigent/onboarding/harness_auth.py: store a key /
gateway (secret → keychain, else ~/.omnigent/secrets.json; a providers: entry
referencing keychain:<name>, never the raw key), adopt an existing host env
var by reference (env:<VAR>, value never read), and detect adoptable env
credentials (non-secret descriptors only). First provider on a family becomes
the default; unsupported families/kinds are refused.
- New host.store_secret / _result frame pair; host daemon handler resolves the
harness→family, calls the core, and re-reports readiness so the badge flips
without a reconnect. Pi maps to its preferred anthropic family.
- New route POST /v1/hosts/{id}/harnesses/{harness}/credential (owner-scoped,
allowlisted, flag-gated) + registry pending_secret_writes plumbing + tunnel
result resolution.
- Regenerated openapi.json.
Tests: core unit tests (incl. the no-raw-secret-in-config invariant), frame
round-trip + telemetry-redaction, host-handler unit tests, and a full route
integration test over a fake tunnel (ownership, flag-off, allowlist, failure
mapping). 243 pass across the affected suites.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* feat(onboarding): detect adoptable credentials on the host (adopt flow)
Adds the read side of the adopt flow: a host.detect_credentials frame pair +
GET /v1/hosts/{id}/credentials/detected that reports the credentials already
present on the host as NON-secret descriptors (family + source label + env var
name), so the UI can offer a one-click "adopt" instead of asking the user to
paste a key they already have. The value is never read or sent — adopt writes
an env:<VAR> reference via the existing store_secret path.
Owner-scoped + flag-gated like the credential-write route. Decode drops
malformed entries so a garbled payload can't inject a non-string field the UI
would trust. Adds frame round-trip (+ malformed-drop), host-handler, and route
integration (+ flag-off) tests; regenerated openapi.json.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(onboarding): tighten the credential route + adopt guard (Polly review)
Two review fixes on the credential-write path:
- The route gated on ui_installable_harnesses(), which includes the env-auth
opencode/qwen — the host handler then rejected them, turning a client/allowlist
problem into a confusing 502. Add ui_credential_configurable_harnesses() (the
Claude/Codex/Pi families the host can actually write) and gate on it, so
opencode/qwen get a clean 400 with no frame forwarded.
- adopt_env_credential now refuses an env var that isn't set on the host —
adopting an unset var would persist a provider entry that resolves to nothing
at the first turn. (Runs on the runner, so os.environ is the host's env.)
Tests: opencode/qwen added to the 400-rejection parametrize; an unset-env-var
adopt rejection case.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(server): serialize concurrent credential writes to one host (Polly review)
Polly non-blocking note: unlike the install route (which coalesces via
inflight_installs), the credential route had no guard against overlapping
writes. The daemon's write is a non-atomic load→merge→save of config.yaml
(twice — entry, then default), so two writes to one host in quick succession
(a double-click, or key + gateway) could interleave and clobber a sibling
providers: entry.
Add a per-connection credential_write_lock held around the store-secret
round-trip so writes to one host serialize. A gateway/local host still
processes different hosts concurrently (the lock is per HostConnection).
Adds an integration test that holds the first reply and asserts the second
frame only reaches the host after the first completes.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(onboarding): make Pi's auth step UI-authable and trackable
Pi's setup_steps auth descriptor was still the M1 shape (action="setup",
command="omnigent setup", status_key=None). Two consequences surfaced in
manual testing: (1) status_key=None made the step "unknown", so the setup
dialog dropped it and wrongly showed "Pi is ready" with no action even though
readiness reported needs-auth; (2) even rendered it was a CLI signpost, not
the credential form.
Pi is UI-authable now (PR A gave it the needs-auth readiness axis; the UI has
the credential form), so its auth step becomes action="auth" (opens the inline
form, keyed on kind=="auth"), command=None (Pi has no subscription CLI login),
status_key="authed" (trackable, so it's not dropped and the dialog reflects
the real state). Qwen stays the untracked env-auth signpost (not UI-authable).
Updates the pi test and adds a qwen-stays-signpost test.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* chore: use the `omni` CLI alias (omni setup) in setup guidance
Rename user-facing "omnigent setup" → "omni setup" in the harness setup-step
descriptors, the setup hint, and their doc-comments. `omni` is the installed
console entry point (pyproject: omni = omnigent.cli:main) and is already used
elsewhere in the codebase, so the shorter alias is correct and consistent.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* test: fix CI drift on the M3 backend branch (omni setup + auth action)
Two "Pytest (misc)" failures on this branch were stale test expectations, not
product bugs:
- tests/host/test_connect.py asserted the unconfigured-launch error names
"omnigent setup", but the earlier `omni` CLI-alias rename made the runtime
message say "omni setup". Update the positive assertion and the cursor
test's negative assertion (which guards that Cursor points at its own
installer, not the generic setup command) to the new spelling.
- tests/test_harness_capabilities.py restricted setup-step actions to
("install", "command", "setup"), but Pi's UI-authable step uses action
"auth" (added when Pi's credential step became a form). Add "auth" to the
allowed set; codex's own two-step assertion is unchanged.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* test: harden the install-flow e2e against a slow picker render
test_install_button_installs_missing_harness opened the agent picker and
immediately clicked the Codex row, but the picker mounts its rows only after
the /v1/agents fetch resolves. Under CI load that render lags, and a menu
opened before the data lands can render empty or re-close on the update — so
the bare open-then-click flaked with a 30s click timeout, the Codex row never
becoming actionable (seen across two different shards). Open the picker, wait
for the Codex row and re-open if the menu flapped, then click. No product
change; passes locally unchanged (the retry is a no-op on the fast path).
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* test: settle agent data before opening the picker in the install e2e
The install-flow e2e flaked (30s click timeout on the Codex row, then on the
picker trigger via an overlay pointer-interception when reopened). Root cause:
the picker opened before the /v1/agents fetch settled, racing the menu-open
against a re-render. Wait for the composer's "Set up Codex" notice (rendered
only once the Codex agent + its unconfigured host state load) BEFORE opening the
picker, then open once and click. Passes locally repeatedly.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* test: stop driving the agent picker in the install e2e (kill the flake)
The picker interaction was redundant — the single seeded Codex agent is already
auto-selected, so the composer's "Set up Codex" notice is present without
opening the dropdown. Driving the picker only added a menu-open-vs-async-render
race that flaked under CI load. Wait for the notice directly (generous 60s) and
click it to open the setup dialog.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* test: wait for network idle before asserting the setup notice (install e2e)
The "Set up Codex" notice depends on two async fetches re-rendering the
composer (/v1/agents auto-selecting the agent, /v1/hosts marking its harness
unconfigured). On loaded CI runners that chain lagged past the timeout and the
assertion raced the still-loading landing screen. Wait for network idle and the
host chip (readiness present) before asserting the notice.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* test: drop networkidle wait in install e2e (WS keeps network busy)
wait_for_load_state("networkidle") never fires in this app — the shell holds a
long-lived sessions/updates WebSocket, so the network is never idle. That wait
just burned its timeout and then raced the still-loading landing screen (the
"Set up Codex" notice was intermittently absent on CI). Replace it with plain
element waits (host chip, then the notice) at a generous 60s, matching every
other e2e_ui test. Passes locally repeatedly.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(host): adopt an env credential under its own family, not the harness's
Review (isaac, MAJOR): pi consumes both the anthropic and openai families, so
the UI can offer an OpenAI env var (e.g. $OPENAI_API_KEY) as adoptable for pi.
`_handle_store_secret` derived the family solely from the harness (pi→anthropic)
and passed that to `adopt_env_credential`, so adopting that var wrote an
anthropic-family provider whose api_key_ref is env:OPENAI_API_KEY — mis-routed
to the anthropic endpoint, failing at run time. For the adopt kind, look the env
var up in the host's detected credentials and use its OWN detected family
(falling back to the harness family if absent). Adds a pi-adopts-OpenAI
regression test.
Also carry the install-flow e2e fix onto this branch: explicitly select Codex
in the picker and stub the /v1/sessions?kind=any agent scan so the seeded-DB
agents don't leak in and leave Claude Code selected (a CI-only flake).
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* fix(host): harden the UI credential-write path (review feedback)
Addresses Polly's blocking finding + hardening notes and Pat's nits on the
store_secret/adopt path:
- BLOCKING (adopt boundary): the daemon's adopt handler fell back to the
harness-derived family when an env var wasn't detected, and adopt_env_credential
only checked the var was *set*. An owner hitting the raw API could name any set
env var (a DB password, an unrelated secret) and have it persisted as a provider
credential sent to the vendor endpoint. Now the handler refuses an env_var that
isn't in detect_adoptable_credentials() (no fallback) — enforcing server-side the
same "only adopt what was detected" restriction the UI presents.
- secrets.py: create the file-backend secrets.json 0600 atomically via
os.open(O_CREAT, 0o600) instead of open()+chmod-after, which briefly left a
freshly-created file group/world-readable. Now network-triggerable, so worth
closing. Fixes the stale "0600 from the start" comment.
- adopt_env_credential: presence-only env check (`in os.environ`, not `.get`) so
the "never reads the value" contract stays literally true.
- gateway base_url: reject a non-http(s) scheme at write time rather than writing
a malformed provider entry that fails opaquely at the first turn.
- connect.py: hoist the harness_auth / provider_config imports to module top
(no circular import) to match the sibling onboarding imports.
Adds regression tests: adopt refuses an undetected env var, gateway rejects a
non-http base_url, and secrets.json is 0600 even under a permissive umask.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
---------
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
The Docker entrypoint's build_app() was constructing RuntimeCaps()
bare, so the llm:, policies:, and routing: blocks in a docker
deployment's config.yaml were silently ignored. This meant:
- Builtin policies that read event["llm_client"] (e.g.
deny_trivial_to_expensive_model) would always see None and abstain.
- default_policies declared under policies: would never fire.
- LLM-based and external routing clients were never built.
Mirror the logic from cli.py: parse_server_llm / parse_default_policies
/ routing client construction are now applied before RuntimeCaps is
passed to init_runtime, putting docker deployments on par with the
CLI-started server.
Fixes#3159
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
Follow-up to #3189. Every session serialization path collapses per-user
`omnigent.pinned.<user>` keys via `_labels_for_viewer` except
`_child_session_summary_from_conversation`, which passed `conv.labels` through
raw. Child sessions aren't pinnable today (the pin affordance lives only on
top-level sidebar rows), so this is a latent gap rather than a live leak — but
if a shared child were ever pinned, its summary would expose another viewer's
pin key.
- Strip any `omnigent.pinned.<user>` key from a child summary's labels. No
collapse-to-canonical: there's no pin to surface, just the defensive strip.
- Test: a child carrying two users' pin keys yields a summary with no pin key,
while unrelated labels survive.
- Correct the stale `useMigrateLocalPinsToServer` docstring: the migration
patches the pinned-list cache (like `useTogglePinnedConversation`), it does
not invalidate the pinned query.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(pi-executor): add supportsUsageInStreaming:false to databricks-completions
Gemini, Qwen, inkling, and other non-OpenAI models in the databricks-completions
provider reject stream_options (which Pi sends with include_usage:true by default)
with 400 'unknown field'. Add supportsUsageInStreaming:false to suppress it,
matching what pi_native_credentials.py already does for omnigent-completions.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi-executor): use openai-responses for newer GPT models (gpt-5-5, gpt-5-6-*)
Newer GPT models reject function tool calls via /chat/completions with 400.
The Databricks Responses API (/ai-gateway/codex/v1/responses) now supports
tool-result chaining on subsequent turns (previously it did not).
- Add databricks-openai provider using openai-responses at /ai-gateway/codex/v1
for gpt-5-5, gpt-5-6-*, gpt-5-3-codex (matches pi_native_credentials routing)
- Keep databricks provider (openai-completions at /serving-endpoints) for
older GPT models (gpt-5-4, gpt-5-4-mini) that work fine with /chat/completions
- Add _pi_needs_responses_api() helper mirroring pi_native_credentials
- Update _pi_provider_for_model() to route to databricks-openai when needed
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi-executor): add kimi to reasoning model fragments
kimi-k2-7-code streams output on reasoning_content channel like GLM/DeepSeek.
Without reasoning:true in the model entry Pi ignores reasoning_content and
sees an empty stream, throwing 'Stream ended without finish_reason'.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(test): update kimi model entry to expect reasoning:true flag
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi-native): add reasoning:true to kimi/glm/deepseek model entries
These models stream output on reasoning_content channel. Pi's openai-completions
parser requires reasoning:true on the model entry to consume that channel;
without it the stream has no content and the turn ends with
'Stream ended without finish_reason'.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(pi-native): exclude qwen3 from completions provider
qwen3 models return content as a typed array [{type:'reasoning',...},{type:'text',...}]
when tools are present, causing Pi's streaming handler to produce [object Object].
Same root cause as gpt-oss; same fix.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix: add inkling to reasoning model fragments and LLM detection
Both kimi and inkling stream output on reasoning_content channel with
content=null. Added inkling to _PI_REASONING_MODEL_FRAGMENTS (executor),
reasoning:true model entry condition (pi-native), and LLM name detection
tokens so it appears in the model list.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor: use allowlist for GPT completions-compatible models
Instead of a denylist of specific model ids that need the Responses API,
maintain an allowlist of GPT models known to work with /chat/completions.
Any GPT model not in the allowlist defaults to Responses API — safer
for new models not yet explicitly tested.
The executor's _pi_needs_responses_api now delegates to the same
implementation in pi_native_credentials for a single source of truth.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
Smart routing was gated behind an OMNIGENT_SMART_ROUTING=1 opt-in on top of the
routing/llm config. The env is redundant: build the routing client whenever the
config supplies one — a server llm: block (built-in judge) or a
routing.provider=external block (external routes:select service). Remove the env
gate in cli.py and refresh the stale references in app.py, advise_models.py, and
web capabilities.ts. Server smart_routing_enabled already keyed on the resolved
client, so the /v1/info signal is unchanged.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
The modular-registry proposal described the phases as thin numbered lists.
Turn them into a concrete, verified implementation plan reviewers can cost:
- Add a "Current state (verified 2026-07-24)" subsection grounding the plan
in the tree at main (59e6b70e): data model ready but no native_providers
field; run_<x>_native already near-uniform (only claude/codex/antigravity/
opencode carry extra kwargs); coverage uneven across hubs (resume 10,
chat-redirect 6, interrupt 9, stop 7); dead _HARNESS_MODULES literal still
present; harness_catalog() emits no native-agent rows.
- Phase 1 (core-only seam): 8 PRs (1.1–1.8) in a table with scope, key files,
dependencies, risk, and estimates. 1.1 provider model + resolver is the
additive foundation; 1.5 runner launch/terminal-route is the risk center.
- Phase 2 (community + web): 4 PRs (2.1–2.4).
- Add an effort summary: ~26–37 engineer-days across ~12 PRs, critical path
1.1 → 1.2 → 1.5 → 2.2 → 2.3. Refresh the Bottom line to match.
Docs-only; no code paths affected.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* fix(web): remove "Create new project" from the project picker menu
Projects are created via the + icon next to the Projects header in the
sidebar, so the picker's own "Create new project" row was a redundant,
second entry point. Drop it (and the inline new-project input it toggled)
from ProjectPickerMenu, leaving search, the project list, and "Remove
from <project>".
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* test(e2e_ui): file sessions via the + button after dropping picker create
The project picker no longer offers an inline "Create new project" row, so
the e2e helpers that drove that flow broke. Rewrite `_move_to_new_project`
to create the empty project from the Projects-header + button first, then
file the session via the kebab picker by name.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* feat(sessions): persist pinned sessions server-side as a per-user label
Pins were client-only (localStorage), so they didn't follow a user across
devices. Move them to a server-side per-user session label so a pin persists
and stays per-user even on shared sessions.
- Store: `omnigent.pinned.<user_id>` label (value = epoch-ms pin time);
`pinned_label_key()` hashes over-long user ids to fit the 128-char key
column. `list_conversations(pinned=True, pinned_owner=…)` filters to the
caller's own key.
- Route: `GET /v1/sessions?pinned=true` enumerates the caller's pins
(independent of the loaded window); PATCH rewrites the client's canonical
`omnigent.pinned` to the caller's per-user key, and `_labels_for_viewer`
collapses it back on read so the per-user dimension never crosses the API
and no viewer sees another user's pin key.
- Write-integrity: reject any client-supplied suffixed `omnigent.pinned.<user>`
key so a caller can't pin/unpin for someone else.
- Forks drop per-user pin keys by prefix (a clone must not inherit pins).
- Web: server-authoritative `usePinnedConversations` + optimistic
`useTogglePinnedConversation`; Pinned section ordered by pin timestamp;
one-time localStorage->server migration that retains pins whose write failed.
- Guard `relativeTime`/`absoluteTime` against non-finite input (no more "NaNy").
Co-authored-by: Isaac
* test(e2e-ui): drive visual-snapshot pins via ?pinned=true, not localStorage
The populated-sidebar visual baseline seeded the pinned session in localStorage,
but pins are now server-authoritative (GET /v1/sessions?pinned=true). Under the
new model the localStorage seed is ignored and the bare-list stub answered the
pinned query too, so every row rendered as pinned → baseline mismatch (the
non-blocking UI Snapshot job).
- Split a `?pinned=true` route out from the bare-list regex (which now also
excludes `pinned=`, mirroring the existing `project=` exclusion) and return
just the pinned row, carrying the canonical `omnigent.pinned` label.
- Drop the `omnigent:pinned-conversation-ids` localStorage seed.
- Apply the same fix to the pinned-project flyout baseline (it passed only by
luck — its bare-list stub happened to return exactly the one pinned row) and
give its row the pin label so it's explicit, not incidental.
Co-authored-by: Isaac
* fix(sessions): let read-only collaborators pin a shared session
Pinning moved server-side (per-user `omnigent.pinned.<user>` label) but the
session PATCH gated all label writes at LEVEL_EDIT, so a read-only collaborator
on a shared session could no longer pin it — a regression from the localStorage
model, which had no permission check.
- Gate a pin-only PATCH (labels == {omnigent.pinned}, no other field) at
LEVEL_READ: pinning is a personal per-viewer preference, not an edit to the
session, so anyone who can SEE it may pin it. Any other field keeps the
edit/owner requirement. Unpin ("" value) is still pin-only, so it downgrades
too. The `?pinned=true` list is already scoped `accessible_by`, so a shared
pin surfaces on "Shared with me".
- Tests: a LEVEL_READ grantee can pin AND unpin a shared session; the downgrade
stays narrow (a non-pin label, or a pin bundled with one, still 403s).
- Rework the access-tier comment to match the if/elif/else (READ / OWNER / EDIT).
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Co-authored-by: Isaac
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
PRs #3148 (extract native terminal orchestration) and #3149 (split the
native app-session test monolith) landed the two remaining Phase 0 file
splits. Update the proposal to reflect reality:
- §1 runner hub: app.py is now ~10.1k lines (was ~20.1k) plus the new
omnigent/runner/native/orchestration.py (~6.5k); drop the stale absolute
line-number anchors and clarify that the dispatch arms and interrupt/stop
closures stayed in app.py while the builders/mirrors moved out.
- Phase 0: mark both runner/app.py and the test monolith Done, noting the
single-orchestration.py outcome (vs the proposed three-way split) and the
nine concern-scoped test modules + shared conftest.py.
- Risk section: re-anchor the forwarder registry to _AUTO_FORWARDER_TASKS in
its new home and note the risk now shifts to Phase 1.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
* refactor(server): split sessions.py into domain sub-modules
sessions.py (7799 lines) is split into 8 focused route files under
_sessions/:
routes_core.py — CRUD, list, WS updates, fork, switch-agent
routes_hooks.py — /hooks/* and /policies/evaluate
routes_items.py — /items and /child_sessions
routes_resources.py — /resources/* (terminals, files, environments)
routes_browser.py — /browser/*
routes_elicitations.py — /elicitations/*
routes_events.py — /events, /stream, DELETE /sessions/{id}
routes_permissions.py — /permissions/*, /owner
routes_agent.py — /agent, /agent/contents, /mcp
Each file exports a register_X_routes(router, ...) function.
create_sessions_router() becomes a thin delegator (~533 lines).
helpers.py gains proxy wrappers for _same_provider_family,
_agent_is_native, _agent_carries_native_fork_history,
_presentation_labels_for_agent, and _reset_runner_resources_after_switch
so existing test monkeypatches on sessions.<name> continue to work.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* refactor(server): move sessions/ route sub-modules out of _sessions/
Convert sessions.py to a sessions/ package and move the 8 routes_*.py
files from _sessions/ into sessions/, so the public package layout is:
routes/sessions/__init__.py (facade, formerly sessions.py)
routes/sessions/routes_core.py
routes/sessions/routes_hooks.py
routes/sessions/routes_items.py
routes/sessions/routes_resources.py
routes/sessions/routes_browser.py
routes/sessions/routes_elicitations.py
routes/sessions/routes_events.py
routes/sessions/routes_permissions.py
routes/sessions/routes_agent.py
_sessions/ retains only the private internals (common, helpers,
orchestration) that do not need public names.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(server): use facade indirection for session_stream and get_agent_cache consistently
routes_browser, routes_events, and routes_hooks were still calling
session_stream.publish() and get_agent_cache() via the direct module
binding. Apply the same facade-indirection pattern already used in
routes_core so all call sites are consistent and test monkeypatches on
sessions.session_stream / sessions.get_agent_cache are honored.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(server): fix lint, _to_agent_object closure, and pyproject.toml exemptions
- Move _policy_type, _policy_description, _to_agent_object from inside
register_permissions_routes closure to module-level in routes_permissions.py
so routes_agent.py can import them directly. Fixes NameError crash on
GET /sessions/{id}/agent in server-approvals tests and E2E tests.
- Add missing 'return router' at end of register_permissions_routes (was
missing after the closure reorganization).
- Import the three helpers explicitly in routes_agent.py.
- Update pyproject.toml per-file-ignores to cover sessions/*.py and
sessions/__init__.py with the same exemptions the original sessions.py
had (ARG001, ARG002, BLE001, E501, F401, F403, F405) so pre-commit
ruff passes.
- Run ruff format on all sessions/ sub-modules.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(server): fix all proxy/monkeypatch misses and restore noqa directives
Route sub-modules were importing _X_impl directly instead of using the
facade proxy, causing monkeypatch(sessions, '_X', ...) to have no effect.
Fix by removing all '_X_impl as _X' imports from routes_*.py — the star-
imports from _sessions.helpers and _sessions.orchestration already bring
in the correct facade-delegating proxies.
Additional fixes:
- Access _SESSION_UPDATES_MAX_WATCHED, _SESSION_UPDATES_RESCAN_INTERVAL_S,
_SESSION_UPDATES_HEARTBEAT_INTERVAL_S through the facade in routes_core.py
so monkeypatch(sessions_routes, '_SESSION_UPDATES_*', N) works.
- Use _load_agent_spec_for_session proxy (not impl) in routes_resources.py.
- Access get_caps() through facade in routes_hooks.py evaluate_policy so
monkeypatch('omnigent.server.routes.sessions.get_caps', ...) fires.
- Restore noqa: BLE001 and F401 directives in _sessions/helpers.py and
_sessions/orchestration.py that were stripped by the RUF100 auto-fix.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(server): delete old sessions.py, fix remaining facade proxy misses
- Delete omnigent/server/routes/sessions.py (the file was rm'd in a prior
commit but never staged; CI was still linting it and seeing F403/F405).
- Route _HOST_BOUND_RUNNER_CONNECT_GRACE_S through the facade in
routes_events.py (3 call sites) so monkeypatch(sessions_module,
'_HOST_BOUND_RUNNER_CONNECT_GRACE_S', ...) is honored.
- Route _recover_subagent_status_forward_via_parent through facade
in routes_events.py.
- Route _registered_runner_id through facade in routes_core.py.
- Route _BROWSER_ACTION_AWAIT_S through facade in routes_browser.py.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(server): route patchable names in routes_hooks.py through facade
All five hook handlers and evaluate_policy use module-level timeout
constants (_CLAUDE_NATIVE_PERMISSION_HOOK_TIMEOUT_S etc.) and auth
helpers (_get_user_id, get_caps, get_agent_cache) that tests monkeypatch
on the sessions facade module. Access them through _sf (the facade) at
call time so monkeypatch(sessions_route, '_CLAUDE_NATIVE_PERMISSION_HOOK_TIMEOUT_S', 0.1)
and monkeypatch('omnigent.server.routes.sessions.get_caps', ...) are honored.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
Re-running CI today means pushing an empty commit or rebasing, which fires a
push event and dismisses existing approvals (branch protection keeps
dismiss-stale-reviews on to block approve-then-swap). A `/rerun` comment
re-runs failed jobs on the existing head SHA instead -- no new commit, so
approvals survive.
Authorized to the PR author or a write-access commenter. Only re-runs the
mock-LLM `pull_request` test suites; the merge gates and Polly AI Review are
left alone. Single file (no privileged relay) because issue_comment gets a
writable base-repo token even for fork PRs.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
--edit-last edits the most recent PR comment regardless of author or
content, so it was overwriting the UI preview comment when both workflows
ran on the same PR. Switch to the same find-by-marker + PATCH approach
used by ui-preview.yml so each workflow manages its own comment.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(web): make session rename optimistic so the new name shows instantly
Renaming a session left the stale name in the sidebar for the duration
of the PATCH round-trip: all cache patching happened in the mutation's
onSuccess, so the row only repainted once the server responded.
Move the cache overlay into onMutate so the new title paints on the next
frame, snapshot the old title for rollback, and restore it in onError.
onSuccess still reconciles with the server-confirmed title + updated_at
and keeps the deliberate no-refetch behavior (an immediate GET races the
search-index reindex). Also patch the ["project-sessions", name] caches
that project folders render from — the flat ["conversations"] overlay
never touched them, so a filed session's row stayed stale until the WS
reconcile.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
* fix(web): cancel in-flight list queries before optimistic rename overlay
Close the in-flight-reconcile clobber race flagged in review: an
already-running GET /v1/sessions reconcile poll (or a WS-triggered
fetch) could resolve after onMutate and overwrite the optimistic title
with the stale search-indexed name. Cancel the ["conversations"] and
["project-sessions"] queries in onMutate before overlaying so no
in-flight fetch can win.
Co-authored-by: Isaac
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
---------
Signed-off-by: Serena Ruan <serena.rxy@gmail.com>
Add an opt-in agent_name field to SessionCreatedEvent. Only polly and
debby are populated — all other agent names are withheld to avoid leaking
user-defined agent names in telemetry.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* feat(sandbox): support OMNIGENT_CONTAINER_RUNTIME env var for container runtime selection
Allow switching the container runtime (e.g. from docker to podman) via the
OMNIGENT_CONTAINER_RUNTIME environment variable instead of requiring per-agent
YAML configuration. The per-agent container_runtime key still takes precedence
over the env var.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(test): add missing top-level `Any` import in test_local.py
Ruff flagged F821 (undefined name) because `Any` was used in a
runtime dict annotation but only imported inside a nested function.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(test): read version dynamically in crash handler test
The test hardcoded "0.6.0.dev0" which breaks when the installed
version diverges from the source (e.g. after a version bump).
Read omnigent.version.VERSION at runtime instead.
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* Revert "fix(test): read version dynamically in crash handler test"
This reverts commit 53855f5c10e3573e9d1ddbfd2afb0bd76abbc91e.
* fix: address review comments on container runtime PR
- Make container_runtime field explicitly Optional to avoid misleading
type annotation and unnecessary type-ignore
- Update parser docstring to mention OMNIGENT_CONTAINER_RUNTIME as an
additional default source
- Update shell script header comment to say "container runtime" instead
of "Docker"
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix(test): add autouse fixture to clear OMNIGENT_CONTAINER_RUNTIME
Prevents the host environment from leaking into tests that assume
the default runtime is "docker".
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* style: add missing blank line before autouse fixture
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* fix: address additional review comments on container runtime PR
- Rename _ALLOWED_RUNTIMES to ALLOWED_RUNTIMES (public API used
cross-module by the parser)
- Reject container_runtime: null in YAML instead of silently falling
back to the env var default
- Add test for container_runtime: null rejection
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
---------
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
* feat(onboarding): report the installed-but-unconfigured harness state
Add the credential axis to the picker-facing readiness map so Claude, Codex,
and Pi report the yellow "installed but no credential" state — the signal the
web setup dialog needs to know when to offer an "Add credential" action
(the M2 keystone of Setup-From-the-UI). Purely additive: the values
("needs-auth" / "binary-missing") already exist in HarnessAvailability, and
the launch gate (harness_is_configured) stays binary-only, so a not-yet-authed
harness is never blocked from launching.
- New _family_provider_configured(): whether an omnigent-managed provider
(API key / gateway) serves the harness's family, reading the same config
omni setup's overview does. Subscription-kind is excluded (that lives in the
CLI's own login, judged by harness_cli_logged_in). Local, side-effect free,
never raises (fails to "no credential").
- Claude: ready when a provider is configured OR the CLI subscription login is
present (was CLI-login only — an API-key-only user wrongly showed yellow).
Checks config first to avoid the status subprocess on the common path.
- Pi: gains the axis it lacked entirely (was binary-only → always green once
installed). No CLI login, so binary + provider: installed-but-no-provider is
now "needs-auth".
Codex already had this (unchanged, the template). Qwen/OpenCode env-auth
unchanged.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* docs(onboarding): clarify _family_provider_configured checks entry presence
Polly review nit: the helper returns True when a non-subscription default
provider *entry* exists, not when its secret actually resolves — an entry
pointing at an unset env:/keychain ref still reads configured (matching the
secret-blind omnigent setup overview). Reword the docstring from "usable
credential" to "a default provider entry is present" and note the
secret-blind behavior + why it's safe (launch gate is binary-only; signal
only moves toward green). No behavior change.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* refactor(onboarding): address review nits on readiness detection
- Hoist the provider_config import in `_family_provider_configured` to the
module top (no circular import); update the test monkeypatch targets to the
now-module-bound name.
- Drop the internal milestone label from a test docstring.
Co-authored-by: Isaac
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
---------
Signed-off-by: xq-yin <xiaoqian.yin@databricks.com>
* feat(sandbox): parse and validate sandbox.kubernetes.pvc_mounts
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* feat(sandbox): add pvc_mounts volumes to the runner Pod manifest
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* feat(sandbox): thread pvc_mounts through the kubernetes launcher
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* docs(deploy): document sandbox.kubernetes.pvc_mounts
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* feat(sandbox): fail loud on unknown sandbox.kubernetes keys
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* test(sandbox): lock in pvc_mounts collision-order, null read_only, and claim-reuse semantics
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* test(sandbox): pin the reserved-mount HOME prefix to the launcher's _HOME_DIR
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* refactor(sandbox): reuse shared validators in the pvc_mounts parser
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(sandbox): close pvc_mounts reserved-path gaps from review
Reject mount_paths with exactly two leading slashes — POSIX normpath
preserves them so '//home/omnigent' passed both validation gates while
the kernel collapses '//' to '/' at mount time, shadowing HOME. Add
/opt to the reserved prefixes: the host image's omnigent venv lives at
/opt/venv and was shadowable. Both cases now covered in the fail-loud
parametrization.
Co-authored-by: Isaac
Signed-off-by: Bryan Li <bryan.li@gmail.com>
* fix(sandbox): reject pvc_mounts paths that mount over reserved prefixes
The reserved-path check only caught mount_paths at or under a reserved
prefix, so an ancestor like /home or /var passed validation while
mounting over the HOME emptyDir mountpoint or the Secret projections.
Reject ancestors too, and reserve /run, /var/run, and /var/lock in full
so the Debian image's /var/run -> /run and /var/lock -> /run/lock
symlinks can't alias around the lexical check.
Co-authored-by: Isaac
Signed-off-by: Bryan Li <bryan.li@gmail.com>
---------
Signed-off-by: Bryan Li <bryan.li@gmail.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
## Related issue
Closes F-CR-6
## Summary
- `DeepLink.parse` validated the `/c/<id>` segment with only `!contains("/")`, but Foundation's `URL.path` is percent-DECODED — so `omnigent://host/c/id%3Fview=terminal` exposes `?` as a literal in the path and smuggles a query (and `%23` a fragment, `%2e%2e` a `..`, `%00` a control char) past the intended "/c/<id> only" shape. Added a denylist that rejects `?`, `#`, `/`, `.`, `%`, and control chars in the decoded id, so an encoded separator that `URL.path` decoded into one of those is dropped.
- The denylist deliberately does NOT assume the id's exact format (the server emits 32-hex uuids today, but the SPA's `/c/:id` route accepts any non-slash segment); the SPA stays the authority on id validity, and a future id scheme (ULID, nanoid, base64) won't be silently rejected. Benign non-canonical ids like `conv_abc` are accepted; only structure-smuggling is blocked.
- Documented the custom-scheme hijack risk in `DeepLink.swift`: iOS doesn't verify single ownership of `omnigent://`, so a co-installed app can read the link's host + id (metadata disclosure). For managed Databricks domains that can serve an `apple-app-site-association`, prefer verified Universal Links; the custom scheme is retained for BYO/OSS servers that can't host AASA, with the interception risk documented.
## Test Plan
- Unit tests (`OmnigentTests/DeepLinkTests`): 19 cases, all pass — including `testRejectsSmuggledQueryViaEncodedQuestionMark` (`%3F`→`?`), `testRejectsSmuggledFragmentViaEncodedHash` (`%23`→`#`), `testRejectsEncodedDotAndDotDot` (`%2e%2e`), `testRejectsControlCharacters` (`%00`/`%0A`/`%7F`), `testRejectsMalformedPercentEscape` (`%zz`), and `testAcceptsBenignNonCanonicalIds` (`conv_abc`/`x`/`not-a-uuid` are accepted — no smuggled structure).
- UI tests (`OmnigentUITests`): 6 cases via a DEBUG-only `--omnigent-open-url` launch-argument seam that routes the link through the real `handleDeepLink`/`DeepLink.parse` (XCUITest can't reliably deliver custom-scheme URLs on this toolchain). `testValidDeepLinkShowsConsent` (valid link → consent alert), `testBenignNonCanonicalIdIsAccepted` (`conv_abc` → consent alert), and rejection tests for smuggled `?`/`#`/`..`/control-char (no alert). A `--omnigent-reset-state` flag wipes persisted server state so each case starts with no known server. All pass on the iOS simulator.
- Manual simulator verification: drove `xcrun simctl openurl` against the running app with `OMNIGENT_DEEPLINK_TRACE` set; NSLog trace confirmed `ACCEPTED` for the valid link and `REJECTED` for all 5 smuggling/malformed links (smuggled `?`/`#`, `..`, control char, non-id).
## Demo
N/A
## Type of change
- [x] Bug fix
- [ ] Feature
- [ ] UI / frontend change
- [ ] Refactor / chore
- [ ] Docs
- [x] Test / CI
- [ ] Breaking change
## Test coverage
- [x] Unit tests added / updated
- [x] E2E tests added / updated
- [x] Manual verification completed
- [ ] Existing tests cover this change
- [ ] Not applicable
## Coverage notes
Manually verified end-to-end on the iOS simulator: launched the app with `OMNIGENT_DEEPLINK_TRACE=1` and sent six real `omnigent://` links via `xcrun simctl openurl`. The NSLog trace showed `ACCEPTED` for the valid link and `REJECTED` for all smuggling/malformed links, proving the fix through the real `DeepLink.parse` → `handleDeepLink` path. The DEBUG-only `--omnigent-open-url` / `--omnigent-reset-state` launch-argument seam and `OMNIGENT_DEEPLINK_TRACE` NSLog logging are compiled out of Release builds (gated by `#if DEBUG`), so there is no production behavior change from the test infrastructure.
* fix(cli): normalize Azure Databricks custom-URL workspaces to their canonical host
omni login (and omni host) failed for Azure Databricks workspaces with a custom
(vanity) URL like https://mydomain.azuredatabricks.net/?o=<workspace_id>: the
vanity edge 303-redirects the unauthenticated probe to /login instead of
answering, so _databricks_workspace_login_target does not recognize the
Databricks posture and login fails. The canonical host
adb-{workspace_id}.{workspace_id % 20}.azuredatabricks.net does answer, and the
?o=<workspace_id> selector already carries the id.
Rewrite the custom host to the canonical adb- form in _resolve_server_url (the
shared normalization every --server entry point uses, so omni host is covered
too). Only *.azuredatabricks.net hosts that are not already the adb- form and
carry a numeric ?o= are touched; AWS/GCP hosts, canonical URLs, and URLs without
a selector are left unchanged.
Closes#2781
Signed-off-by: Enes Yilmaz <enesyilmaz5157@gmail.com>
* fix(cli): probe before adopting the canonical Azure Databricks host
The custom-URL fix landed the canonical adb- host rewrite unconditionally in
_resolve_server_url, so a wrong synthesis could strand the user on a host they
never typed, and the unit tests only re-asserted the implementation's own
arithmetic (123 % 20 == 3), which would pass under any modulus.
Try the URL as the user gave it first. Only when that fails to resolve, and only
for an Azure vanity workspace URL carrying a numeric ?o=, synthesize the
canonical host, probe it, and adopt it if it answers. A dead synthesis now falls
back to the user's URL instead of replacing it.
The shard rule remains an observed regularity rather than a documented contract
(Microsoft calls the segment a random number and treats properties.workspaceUrl
from the ARM API as authoritative), so the probe keeps it off the load-bearing
path. Docstrings say so plainly.
Also:
- _canonicalize_azure_databricks_url is now _canonical_azure_databricks_url and
returns None to decline, so a caller can tell "not applicable" from "no change".
- Guard the selector with isascii() as well as isdecimal(): str.isdecimal()
accepts non-ASCII digits that int() also parses, which synthesized a
nonsensical host.
- _probe_root reduces a URL the way _workspace_api_server_url does before
probing. Without it the comparison against the expansion's result never
matched (it drops the ?o= selector first, and that selector is what makes a
URL a candidate), and the new probe requested /?o=123/v1/me.
- Replace the tautological shard assertions with five real observed
workspace/host pairs, and drive the resolver tests through the real expansion
with only httpx scripted, since a stubbed expander cannot catch the above.
Signed-off-by: Enes Yilmaz <enesyilmaz5157@gmail.com>
* docs(cli): drop issue-number refs from Azure canonical-host comments
The repo's comment convention says code comments should describe the
scenario, not reference issue/PR numbers. Remove the (#2781) tags from
the _canonical_azure_databricks_url / _resolve_server_url docstrings and
the vanity-URL fallback test; the surrounding prose already explains the
Azure vanity-host case without needing the external link.
Co-authored-by: Isaac
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
---------
Signed-off-by: Enes Yilmaz <enesyilmaz5157@gmail.com>
Signed-off-by: Pat Sukprasert <pat.sukprasert@databricks.com>
Co-authored-by: Pat Sukprasert <pat.sukprasert@databricks.com>
* fix(runner): resolve and re-materialize file attachments on remote-runner history reload
Signed-off-by: Sunny Yang <sunnyadn1130@gmail.com>
* fix(runner): seed the native-session compaction anchor; tolerate malformed file metadata
Native-harness sessions skip the history reload entirely, which also
skipped seeding the last server item ID that harness compaction
persistence anchors on — compactions then silently stopped persisting.
Session create now fetches just the newest item ID (newest-first, single
item, no attachment downloads) for native harnesses.
A 200 metadata response with an unparseable body no longer aborts
attachment resolution: both resolvers (the runner's message-content
resolver and the claude-native transcript rebuild) fall back to the
content response's Content-Type for the media-type hint.
Signed-off-by: Sunny Yang <sunnyadn1130@gmail.com>
* refactor(attachments): centralize file_id resolution and reference-line emission in native_attachments
The transcript rebuild and the runner each carried a full copy of the
file_id fetch-and-inline pipeline, and nine native executors repeated
the same materialize-or-marker block. Both now live in
native_attachments: resolve_file_id_block() serves the runner and the
transcript rebuild, attachment_reference_line() serves the executors,
and ATTACHMENT_MARKER_STRIP_PATTERN replaces four hand-copied forwarder
regexes. Materialized filenames are sanitized the same way as marker
names so a bracketed filename cannot break the marker consumers, and
the resume dedupe short-circuits on file size before reading bytes.
Signed-off-by: Sunny Yang <sunnyadn1130@gmail.com>
* fix(attachments): replay resolved history attachments as structured content
Cold-started claude-sdk sessions flattened prior turns into a text
prefix, so a resolved historical image reached the model as the marker
[image: name, media_type, N base64 chars]. The bytes never arrived, which
leaves the #882 symptom in place for that harness: the model describes an
attachment it cannot see.
Prior-turn attachments now replay as real Anthropic image/document blocks
via the existing converter, interleaved in transcript order. Text-only
history still takes the plain-string path and renders byte-identically,
unresolved attachments keep their existing marker, and base64 still never
enters prompt text.
Materialization also derives its collision suffix from a content hash
rather than a random one, so a history carrying two distinct uploads of
the same filename keeps one file per payload instead of gaining a copy on
every transcript rebuild.
The two tests that asserted the compact-placeholder shape are replaced by
cold-reload tests: that shape is the behavior being corrected, but the
invariant those tests protected (no base64 in prompt text) is asserted
against the prompt's text blocks.
Signed-off-by: Sunny Yang <sunnyadn1130@gmail.com>
* refactor(attachments): collapse duplicated prompt-shape branches
The structured and plain-text arms of _build_prompt returned the same
value whenever the latest message was multimodal, and re-scanned the
block list to decide which arm to take. Coalescing already leaves an
all-text history as one block, so the block count answers that.
Materialization's second identity check was a no-op guarding a write
that produces the same bytes, so the collision path flattens to one
branch.
Signed-off-by: Sunny Yang <sunnyadn1130@gmail.com>
* refactor(tests): keep runner conftest identical to upstream
Move the file-server fake's items/failure/malformed-meta behaviors out of
the shared _FakeFileServerClient into local subclasses in the one file
that uses them, so conftest.py stays in sync with upstream and per-test
modes stay next to their tests.
Signed-off-by: Sunny Yang <sunnyadn1130@gmail.com>
---------
Signed-off-by: Sunny Yang <sunnyadn1130@gmail.com>
Co-authored-by: Pat Sukprasert <pattara.sk127@gmail.com>
The before-quit handler defers the quit until serverManager.shutdown()
finishes, then re-issues app.quit() as the *only* way the quit ever
proceeds. Re-issuing app.quit() after before-quit's preventDefault() is a
known intermittently-unreliable Electron behavior (electron/electron#4994,
#33643, #39094); when it no-ops, or shutdown hangs (a stuck
'omnigent server stop'), the app stays up with its window still open —
matching 'sometimes the app is still running and refuses to quit'.
- Hard safety cap: app.exit(0) after quitCleanupTimeoutMs (unref'd) if
graceful cleanup + the re-issued quit haven't terminated. Normal cleanup
(<6s) completes well under the 10s cap; it only trips when stuck.
- Evaluate resolvedCliPath() inside an async IIFE so a future throw becomes
a rejection caught by .catch, never stranding the quit.
- Install fallback: when quitAndInstallIfPending() returns true but
quitAndInstall() doesn't actually quit (staged update gone), a short
app.exit(0) fallback still quits.
- unref() the periodic update-check setInterval so it can't keep the event
loop alive at quit.
Adds two regression tests (install-fallback and cleanup-cap) via an
injectable setQuitTimeouts; harness exposes setTimeout/clearTimeout/app.exit.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
The codex-native launch read the spec model only from
executor.config["model"], a key the single-file agent loader never
populates, so a custom agent's declared model: was silently replaced by
the provider default. Read the canonical executor.model first — the same
field the in-process harness and the claude/cursor native launches
consume — and keep config["model"] as a fallback for bundle specs that
pin the model inside the harness config block.
Co-authored-by: Isaac
* fix(loader): reject the bundle type:/config: nesting in single-file executor blocks
A single-file agent YAML written with the bundle config.yaml shape
(executor: {type: omnigent, config: {harness: ...}}) loaded without
complaint: the unknown keys were silently dropped, the declared harness
with them, and a different harness was inferred from the model prefix —
databricks-gpt-* landing on openai-agents instead of the declared
codex-native, with no diagnostics. Reject exactly type:/config: with an
error that shows the flat spelling. Other extra executor keys
(use_responses, extra, ...) keep loading — the compat loader reads them
from the raw YAML.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
* test: spell e2e fixture executors flat instead of the bundle config: nesting
Six runtime-generated single-file agent YAMLs in the e2e/e2e_ui/server
fixtures nested the harness under executor.config — the exact trap the
loader now rejects. They only worked because the dropped harness was
re-inferred from the gpt-* model prefix as the same openai-agents value.
Spell them flat so the declared harness actually flows. The two
spec_version bundle specs (approval agent, elicitation supervisor) keep
the nesting — config.harness is the correct spelling on the strict
parser path.
Co-authored-by: Isaac
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
---------
Signed-off-by: Dhruv Gupta <dhruv.gupta@databricks.com>
The background server spawned by bare `omni` (`_spawn_local_server`)
launched `omnigent.cli server` without `--config`, so the server's
loader returned an empty config and never read `~/.omnigent/config.yaml`.
Its `llm:` (and `policies:`) block was invisible to the detached server,
so self-hosted smart routing silently stayed off (`sys_advise_models` ->
`router_on: false`; `/v1/info` -> `smart_routing_enabled: false`).
Forward `--config <global_config_path()>` when the file exists. Same bug
class as #2386/#2763 (Docker entrypoint dropped `policies:`); this is the
local-spawn instance.
Co-authored-by: Isaac
Signed-off-by: Pranav Setlur <psetlur@gmail.com>
PR #3148 extracted _session_labels_for_runner_spawn into
omnigent.runner.native.orchestration, but _SESSION_STREAM_HEARTBEAT_S
and the stream loop that reads it remained in omnigent.runner.app.
test_session_stream_emits_heartbeat_on_idle located the module to patch
via _session_labels_for_runner_spawn.__module__, which now resolves to
omnigent.runner.native.orchestration — a module that has no
_SESSION_STREAM_HEARTBEAT_S attribute — so the test raised
AttributeError and failed CI on main.
Patch omnigent.runner.app directly, which is where the heartbeat cadence
constant and its consumer actually live.
Signed-off-by: Zeyi (Rice) Fan <zeyi.f@databricks.com>
When the launching process sets CLAUDE_CODE_USE_GATEWAY=1, that
gateway-aware mode keeps tool search enabled so MCP schemas load on
demand. Setting CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS alongside it
would override that mode, disabling all betas and inflating startup
token usage.
Only set CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS when gateway-aware
mode was NOT selected.
Ported from databricks-eng/universe#2298829.
Co-authored-by: harry-yao_data <harry.yao@databricks.com>
Introduce no-op extension points on the conversation store so a subclass
can transform conversation_items.data and control search_text, without
changing OSS behavior:
- _encode_item_data(data_json): identity by default; append's data write is
routed through it so a subclass may compress or encrypt the payload.
- _decode_item_data_batch(stored_list): identity by default; the read paths
(list_items, list_latest_message_items_for_conversations, the FTS-ranked
read) decode a whole page of rows through it before building entities, and
_to_item now takes the already-decoded data. Making the read seam a batch
(not a per-row hook) lets a subclass decode a page in one pass — e.g. a
single bulk decrypt — instead of once per row.
- _item_search_text(item): extracts the search text as before by default;
may return None to skip persisting search_text (and its FTS row) on a
schema that omits the column.
Every default preserves current behavior exactly: the column stays plaintext
Text, and search/FTS are unchanged. This lets a downstream store (Databricks'
MySQL-homed conversation store) envelope-encrypt item payloads at the column
boundary while reusing append/list_items unchanged.
Co-authored-by: Isaac
Signed-off-by: aravind-segu <aravind.segu@databricks.com>
* docs(deploy): correct docker admin bootstrap flow (no auto-generated password)
Signed-off-by: Sai Asish Y <say.apm35@gmail.com>
* docs(deploy): correct remaining generated-password and /data-persistence claims
Signed-off-by: Sai Asish Y <say.apm35@gmail.com>
* docs(deploy): scrub generated-password flow from remaining platform guides
The Docker docs were corrected earlier, but fly / railway / render / modal /
hf-spaces still told operators to read a generated admin password out of the
logs / /data/admin-credentials — a flow that no longer exists (bootstrap never
auto-generates a password; the first admin is claimed via the web Create-admin
form or a pre-seeded OMNIGENT_ACCOUNTS_INIT_ADMIN_PASSWORD).
- Rewrite the first-admin step in each guide to the real flow, and drop the
fake "Created initial admin ... password: <generated>" log block.
- Add a first-visitor security note (unauthenticated /auth/setup while no
password-bearing account exists) to every public-facing guide; fold it into
hf-spaces' "make the Space Public" step where the exposure is most direct.
- render: correct the disk bullet (hashes live in Postgres, not on /data) and
the render.yaml comment that called the anchor path a password file.
Co-authored-by: Isaac <isaac@example.com>
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
---------
Signed-off-by: Sai Asish Y <say.apm35@gmail.com>
Signed-off-by: Pat Sukprasert <pattara.sk127@gmail.com>
Co-authored-by: Pat Sukprasert <pattara.sk127@gmail.com>
Co-authored-by: Isaac <isaac@example.com>
* feat(cli): add `omnigent session import` (inverse of session export)
`session export` writes a portable JSONL but there was no way to load it
back — inspecting a shared/exported session meant hand-writing items into
the store. Add `session import` to close the round-trip: it reads the
session_meta + item lines and recreates the conversation on the target
server as a new session (fresh id each time) via POST /v1/sessions with
the history passed as initial_items.
Details:
- De-aliases the `model` serialization alias back to `agent` per item and
validates each with parse_item_data() client-side before the request.
- Agent binding: reuse the exported agent_id when it exists on the target
server; else fall back to the built-in native agent for the export's
harness (mirrors /v1/imports); else fail with a clear message.
- Creates history-only (host_type=external, no host_id) so no runner
launches. Carries over title/workspace/harness/model/effort overrides.
Known limitation (documented in --help): the server seeds initial_items
under a single synthetic response_id, so exact per-turn grouping is not
preserved. Fine for viewing/debugging; a follow-up server route could
preserve it if needed.
Verified end-to-end: imported the real 260-item export, re-exported, and
diffed — identical item counts and types, agent bound, model<->agent
alias round-trips.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
* fix(cli): scope agent→model de-alias to alias-bearing item types
Polly review caught that the import de-alias applied `model`→`agent` to
every item type, corrupting the two types where `model` is a genuine
field: `compaction.model` (silently dropped) and `routing_decision.model`
(required + collides with its own `agent` field → hard import failure for
any smart-routed session).
Derive the alias-bearing types from the data-model field definitions
(serialization_alias == "model") so the reverse map only fires for
message/function_call/reasoning/slash_command and can't drift. Add
regression tests for compaction and routing_decision.
Also address non-blocking review notes:
- Wrap non-404 create errors in a clean ClickException instead of a raw
httpx traceback.
- Document created_by re-attribution in --help alongside the response_id
caveat.
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
---------
Signed-off-by: Tomu Hirata <tomu.hirata@gmail.com>
2026-07-23 14:25:36 +00:00
1298 changed files with 163072 additions and 84697 deletions
description: Run the Omnigent load test and produce a results file explaining the latencies. Load when the user wants to load-test / stress-test / benchmark Omnigent under concurrency ("load test omnigent", "stress test the server", "how many hosts/sessions/turns can it handle", "load test real agent turns / conversations", "run a load test"). The test makes each simulated user a real omnigent host that creates host-bound sessions and drives real multi-turn conversations with a mocked LLM; it boots its own local stack (dev/loadtest/run.py). Gather inputs, run it, then read the generated summary.md and explain the latency distribution (avg/median/p95/p99, throughput, failures). NOT for single-request latency micro-benchmarks (that is dev/benchmarks/).
---
# Run the Omnigent load test
Drives `dev/loadtest/` end to end: collect inputs → run → read `summary.md` →
explain the latencies. **Each Locust user is a real `omnigent host`** that
registers over the host tunnel, creates host-bound sessions, and drives **real
multi-turn conversations** — every turn is a genuine post→idle loop through the
host's runner, with the **LLM mocked** (zero latency) so the numbers are
Omnigent's own overhead. `-u N` scales the number of hosts.
It **boots its own local stack** (server + mock LLM), so there is no server to
point at, and it runs **from a repo checkout** only. For single-request latency
micro-benchmarks (not concurrency), that is a different tool: `dev/benchmarks/`.
run:uv run --no-project --python 3.12 --with packaging python scripts/update_versions.py check
@@ -139,6 +144,6 @@ jobs:
--title "Bump version to ${resolved}" \
--body "Automated version bump via \`.github/workflows/bump-version.yml\` (mode: \`${MODE}\`, input: \`${NEW_VERSION}\`).
Rewrote \`[project].version\` and sibling \`==\` pins across all three packages (\`pyproject.toml\`, \`sdks/python-client\`, \`sdks/ui\`), the runtime \`VERSION\` constant in \`omnigent/version.py\`, and regenerated \`uv.lock\`.
Rewrote \`[project].version\` and sibling \`==\` pins across all four packages (\`pyproject.toml\`, \`sdks/python-client\`, \`sdks/ui\`, \`integrations/slack\`), the runtime \`VERSION\` constant in \`omnigent/version.py\`, and regenerated \`uv.lock\`.
Generated by \`scripts/update_versions.py\`. Opened via the omnigent-ci App when configured (CI runs automatically); on the GITHUB_TOKEN fallback, re-open or push to kick CI."
body="$(printf 'Drafts a feature-blog post for **%s**, selected by `feature-blog-scout` at the %s release cut.\n\n> **This is a DRAFT.** Before merging, a human must: record the mandatory demo (replace the `DEMO REQUIRED` marker) and do a final voice pass. The hero image and `author: omnigent` byline are auto-generated — review and optionally replace them.\n\n%s\n\nSource release: %s%s\n<sub>Generated by omnigent `.github/workflows/feature-blog.yml`. Review for accuracy before merging.</sub>' "$title" "$TAG" "$summary" "$TAG" "$mention")"
# Reference table of the contributing PRs and whether each already
# ships a demo video (built in the Draft posts step). Reviewers can pull
# an existing recording from a ✅ PR to replace the `DEMO REQUIRED`
# marker instead of re-recording. Omitted if the table wasn't produced.
demo_table=""
if [ -f "/tmp/demo_table_${idx}.md" ]; then
demo_table="$(printf '\n\n### Source PRs — demo videos\nCheck a ✅ PR for a recording you can drop into the `DEMO REQUIRED` marker.\n\n%s\n' "$(cat "/tmp/demo_table_${idx}.md")")"
fi
body="$(printf 'Drafts a feature-blog post for **%s**, selected by `feature-blog-scout` at the %s release cut.\n\n> **This is a DRAFT.** Before merging, a human must: record the mandatory demo (replace the `DEMO REQUIRED` marker) and do a final voice pass. The hero image and `author: omnigent` byline are auto-generated — review and optionally replace them.\n\n%s%s\n\nSource release: %s%s\n<sub>Generated by omnigent `.github/workflows/feature-blog.yml`. Review for accuracy before merging.</sub>' "$title" "$TAG" "$summary" "$demo_table" "$TAG" "$mention")"
existing="$(gh pr list --repo "$SITE_REPO" --head "$branch" --state open --json url --jq '.[].url' | head -n1)"
# An already-open PR just picks up the force-pushed update.
if [ -n "$(gh pr list --head "$BRANCH" --state open --json number --jq '.[0].number // empty')" ]; then
@@ -126,7 +150,7 @@ jobs:
# exempts gh from `set -e`, so a non-zero exit hits the else branch.)
if gh pr create --base main --head "$BRANCH" \
--title "chore(oss): regenerate public lockfiles against public PyPI/npm" \
--body "Automated: regenerated uv.lock + web/package-lock.json against public PyPI/npm, validated by a Docker build + omnigent --help smoke (run ${{ github.run_id }}). Merge to keep the public lockfiles current and buildable."; then
--body "Automated: the lockfiles were out of sync with the manifests, so uv.lock + pnpm-lock.yaml were regenerated against public PyPI/npm and validated by a Docker build + omnigent --help smoke (run ${{ github.run_id }}). Merge to keep the public lockfiles consistent and buildable."; then
echo "Opened the regen PR."
else
echo "::warning::Could not open the regen PR automatically (the GITHUB_TOKEN may be disallowed from creating PRs). The branch '$BRANCH' is pushed with the regenerated lockfiles — open the PR by hand:"
body="$(printf 'Publishes the **%s** release post at `/releases/%s` — the curated GitHub Release notes reformatted into the site'"'"'s narrative, prose-driven style.\n\nGenerated by omnigent `.github/workflows/publish-changelog.yml`. Edit the GitHub Release, not this file.' "$TAG" "$VERSION")"
# Append the demo-video reference table: which feature PRs already ship a
# recording a reviewer can drop into the post'"'"'s `TODO` demo placeholders.
if [ -s /tmp/demo_table.md ]; then
body="$(printf '%s\n\n### Source PRs — demo videos\nCheck a ✅ PR for a recording to replace a `TODO` demo placeholder in the post.\n\n%s' "$body" "$(cat /tmp/demo_table.md)")"
case "$VERSION" in *a[0-9]*|*b[0-9]*|*rc[0-9]*) prerelease=true ;; esac
case "$VERSION" in *rc[0-9]*) prerelease=true ;; esac
{
echo "version=${VERSION}"
echo "tag=v${VERSION}"
@@ -164,7 +164,7 @@ jobs:
echo "Tag ${TAG} already at the converged release commit ${base_sha} — nothing to do." \
| tee -a "$GITHUB_STEP_SUMMARY"
else
echo "::error::Tag ${TAG} already exists at ${tag_sha} (stamped version: ${stamped:-unknown}), which is not the converged branch head ${base_sha}. Delete the tag first if this is recovery (see RELEASING.md)."
echo "::error::Tag ${TAG} already exists at ${tag_sha} (stamped version: ${stamped:-unknown}), which is not the converged branch head ${base_sha}. Delete the tag first if this is recovery (see the maintainer release runbook)."
exit 1
fi
fi
@@ -600,6 +600,10 @@ jobs:
uv run --no-project --python 3.12 --with packaging \
echo "2. Validate the rc from PyPI (see RELEASING.md). The GitHub draft for ${TAG} stays unpublished."
echo "2. Validate the rc from PyPI (see the maintainer release runbook). No GitHub release is created for rc tags (rcs live on PyPI only) — skip straight to the next rc or the final cut."
else
echo "2. Merge the CHANGELOG PR, curate the ${TAG} draft notes, then dispatch finalize-release.yml (tag=${TAG})."
fi
@@ -657,7 +666,6 @@ jobs:
# never re-freezes and doc-sync keeps deriving the right X.Y-docs branch.
--body "🔁 \`/rerun\`: no failed CI runs on the current head (\`${SHA:0:7}\`) to re-run. If a check is stuck *pending*, it needs a push or a maintainer, not a re-run."
exit 0
fi
RERAN=""
while IFS=$'\t' read -r id name; do
[ -n "$id" ] || continue
echo "• Re-running failed jobs in '$name' (run $id)"
# --failed: re-run only the failed jobs (cheapest path for a flake).
# --repo is REQUIRED: this job has no checkout, so `gh run rerun`
# cannot infer the repo from a git remote and would fail client-side.
if gh run rerun "$id" --repo "$REPO" --failed; then
RERAN="$RERAN"$'\n'"- $name"
else
echo "::warning::Could not re-run '$name' (run $id) -- may be in progress."
RERAN="$RERAN"$'\n'"- $name ⚠️ (skipped: already running or not re-runnable)"
fi
done < <(printf '%s\n' "${FAILED[@]}")
NOTE="The \`Merge Ready\` gate re-evaluates automatically when these complete."
gh pr comment "$PR_NUMBER" --repo "$REPO" \
--body "🔁 \`/rerun\`: re-running failed jobs on \`${SHA:0:7}\`:${RERAN}"$'\n\n'"$NOTE"
if [ -n "$(gh pr list --repo "$TAP_REPO" --head "$BRANCH" --state open --json number --jq '.[].number')" ]; then
echo "Bump PR already open for ${BRANCH} — force-push updated it." | tee -a "$GITHUB_STEP_SUMMARY"
exit 0
fi
body="$(printf 'Bumps the omnigent formula to **%s** (new sdist url/sha256, resources regenerated via `brew update-python-resources`).\n\ntest-bot builds the bottles on this PR. Review the resource diff — especially that the extras'"'"' deps survived — then apply the `pr-pull` label to publish bottles and merge.\n\nOpened by omnigent `.github/workflows/update-homebrew.yml`.' "$VERSION")"
gh pr create \
--repo "$TAP_REPO" \
--base main \
--head "$BRANCH" \
--title "omnigent ${VERSION}" \
--body "$body"
echo "Opened tap bump PR for omnigent ${VERSION}." | tee -a "$GITHUB_STEP_SUMMARY"
@@ -5,12 +5,476 @@ generated at release time from each PR's `## Changelog` section, tagged by the
PR's `Type of change` (e.g. `[UI]`); the concise, curated highlights live on the
website under `/releases`.
## [Unreleased]
## [v0.8.1] — 2026-08-03
### Features
- [UI] Reverted the v0.8.0 "Chat/Terminal switcher in the header" change; the
switcher returns to its previous location. (#3931)
- [UI / Feature] Added a Nord color theme (arctic frost-blue palette) to the Appearance settings palette picker.
- [Feature] Per-harness startup command/args overrides via a polymorphic `harness:` key in `config.yaml`. The `harness:` key now accepts a mapping with a `default` plus per-harness `command`/`args` overrides (e.g. `harness: {default: claude-sdk, codex: {command: /usr/local/bin/codex, args: [--config, approval_policy=on-request]}}`). The legacy scalar form (`harness: claude-sdk`) still works and auto-migrates to the mapping form on the next config write. Harness binary-path precedence: `OMNIGENT_<NAME>_PATH` env var > config `harness.<id>.command` > built-in default; `args` follow the same precedence with config `args` as the base and CLI pass-through args appended. The `OMNIGENT_<NAME>_PATH` env var (base id, `-native` suffix stripped) is the canonical per-binary override, standardizing the headless `HARNESS_<NAME>_PATH` and native `OMNIGENT_*_PATH` conventions into one namespaced name; the legacy `HARNESS_<NAME>_PATH` is still read as a deprecated fallback that logs a one-time warning + a CLI startup notice, and is slated for removal in v0.8.0. The pre-existing `omnigent claude --command` flag is deprecated (warns on use, pointing to `OMNIGENT_CLAUDE_PATH`/config) and will be removed in a future release; no other native command gained a `--command` flag — override via env or config.
## [v0.8.0] — 2026-08-03
- [Bug fix] Cursor YOLO sessions no longer stall piloted parents on mirrored tool-approval cards when Cursor leaves a lingering pending gate. (#2338)
- [Feature] codex-native startup-timeout errors now name the resolved provider/model routing (and the login-fallback case) instead of pointing at the runner log (#2843)
- [UI / Bug fix / Feature] Pi can now create task plans and display them in the shared Tasks panel without requiring an optional Pi extension. (#2884)
- [Feature] `detect_loop` builtin policy catches agents stuck retrying the same tool call and prompts for approval to break the loop (#3158)
- [Feature] New `detect_thrashing` builtin policy detects when an agent is stuck in a failure loop and alerts the user to intervene. (#3160)
- [Bug fix] kimi sub-agents now report completion to a parent orchestrator instead of leaving the fan-out waiting forever. (#3166)
- [UI / Bug fix / Chore / Test/CI] Conversations show their newest messages immediately, pin the latest turn below the header, and load older context smoothly near the top. (#3228)
- [Docs / Chore / Breaking] The `run_<x>_native` launchers accept a uniform `extra_args`; the per-harness `<x>_args` keyword is deprecated and will be removed in 0.9.0. (#3244)
- [Bug fix] Intelligent routing no longer drops the first message on a new claude-native session (#3257)
- [Feature] `linux_bwrap` sandboxes and their egress rules now run on the Databricks Lakebox backend. (#3258)
- [Feature] `sandbox.kubernetes.secret_mounts` projects a Secret as a rotation-friendly read-only file volume on the runner (#3280)
- [Bug fix] Session cost panel now shows a per-model breakdown for sub-agents/second heads (e.g. Debby's GPT head, Polly's codex sub-agents) running on an unpinned codex model, instead of folding their usage into the total with no per-model entry. (#3287)
- [Bug fix] Fix `antigravity-native` readiness detection for `agy` CLI installs on macOS where OAuth credentials live in Keychain (#3289)
- [Bug fix] Fixed a race condition where deleting two admins at nearly the same time could leave a deployment with zero admins and no way to recover through the API. (#3304)
- [Docs / Chore] Resuming a goose / hermes / antigravity / qwen / opencode native session now attaches to the live TUI instead of double-posting each message through the Omnigent REPL. (#3314)
- [UI / Feature] Workspace rail now shows files and shells as editor-style tabs, opens shells in-rail instead of replacing the chat, adds a full-screen toggle, and adds a "+" menu for creating shells with a remembered type picker (#3333)
- [UI / Bug fix / Feature / Test/CI] `omnigent setup` can import OpenClaw/acpx coding agents from an auto-detected or user-selected config into the generic ACP harness picker, and `omnigent run --from-openclaw <agent>` can try one without saving it. (#3354)
- [Feature] Polly can launch supported Claude and Codex implementation children in goal mode. (#3362)
- [Bug fix] Claude sessions started with a model alias (e.g. Opus) no longer fail on gateway setups that can't pin the alias — the launch resolves to a routable model id. (#3378)
- [Feature] `omni setup` now supports signing in to Antigravity with Google OAuth through `agy`, alongside Gemini API keys. (#3391)
- [UI / Feature] Archived sessions in Settings are now grouped by date (Today, Yesterday, Previous 7/30 days, month/year) for easier browsing. (#3394)
- [UI / Bug fix] Subagent graph view nodes are now clickable and navigate to the selected session. (#3395)
- [Bug fix] Dedupe codex-native's per-session plugin cache to reclaim disk (#3401)
- [Bug fix / Test/CI] Shared-session sub-agent completion notices are attributed to the collaborator who dispatched the sub-agent. (#3409)
- [Bug fix / Test/CI] Codex sessions keep their selected permission mode when resumed on a replacement host. (#3411)
- [Bug fix] Only session owners can approve tools that run with owner credentials in shared sessions. (#3416)
- [Bug fix] Fixed a bug where a malformed or unrecognized tool-call payload on certain harnesses (notably OpenCode) could silently bypass a configured tool-call policy instead of being blocked or asked. (#3418)
- [Bug fix / Feature / Docs / Test/CI] ACP agents can opt out of Omnigent’s MCP relay with `omnigent_mcp: false`, enabling compatible OpenClaw Gateway ACP registrations. (#3420)
- [Feature / Docs] Shared-session agents can distinguish who wrote each message without changing whose credentials execute the session; operators can hide model-visible author labels with an environment flag. (#3422)
- [Bug fix] `/model` and the startup header no longer name an Omnigent provider and model for ACP-backed sessions, which run on the agent's own auth and model. (#3431)
- [Bug fix] Session tokens saved by `omnigent login` are now created owner-only, so a JWT is never briefly world-readable on first login, and an interrupted write no longer discards every stored token. (#3441)
- [Feature / Docs / Chore / Test/CI] Model selection can now represent provider choices through stable intents and normalized capability metadata. (#3443)
- [UI / Feature] Session owners can grant trusted collaborators permission to approve privileged actions without transferring ownership; ordinary editors can reject but cannot approve. (#3446)
- [Bug fix / Feature / Docs / Chore / Test/CI] Default model selection now follows the active provider catalog instead of release-specific model names baked into runtime harnesses, while retaining general-purpose selection policy and Databricks gateway routability. In cold-cache Databricks environments without GitHub egress, configure `executor.model` or a provider `models.default`. (#3448)
- [Feature / Docs / Chore / Test/CI] Smart routing now chooses only from models discovered on the active runner instead of falling back to release-specific model names. (#3450)
- [Feature / Docs / Chore / Test/CI] The Kiro model picker now follows the models and metadata reported by the installed Kiro CLI. (#3452)
- [Feature / Docs] Minimal `omnigent run` agents now discover their default model instead of using a release-specific built-in endpoint. (#3455)
- [Feature / Docs / Test/CI] Provider setup and unconfigured provider runtimes now select current catalog models instead of release-specific built-in defaults, and fail with explicit configuration guidance when discovery is unavailable. (#3456)
- [Docs / Chore] The Kimi launcher example now respects the default model configured in Kimi Code. (#3457)
- [Bug fix] `omni resume` now accepts a conversation id pasted with stray punctuation (trailing period, quotes, backticks) instead of crashing (#3465)
- [UI / Bug fix] Sidebar New session, Automations, and Inbox icons now sit on the same left column (#3468)
- [Chore] Android app now targets Android 16 (API 36) to stay compliant with Google Play's (#3470)
- [Feature / Test/CI] Nightly prerelease builds: every night a `X.Y.Z.devYYYYMMDD` tag of main is cut automatically; install or update with `scripts/update_nightly.sh` or `omni upgrade --nightly` (#3475)
- [Bug fix / Breaking] Agent CLIs (qwen, goose, kimi, hermes, and generic ACP agents) no longer receive unrelated host credentials such as cloud tokens and other providers' API keys; an agent that authenticates from a variable outside its own family now declares it in `os_env.sandbox.env_passthrough`. (#3479)
- [UI / Bug fix] Conversation sidebar text now scales with the Interface font size setting (#3480)
- [Bug fix] Fixed two `claude-sdk` steering bugs: messages sent during an active turn were answered one turn late (a permanent chat desync), and steering several messages at once dropped all but the last. Steered messages are now buffered correctly and all of them reach the model. (#3484)
- [Bug fix] Fix the first message being silently dropped when a session resumes on a managed sandbox / lakebox (#3488)
- [UI / Bug fix] The workspace file view keeps its scroll position — both the file tree and the open file — when you switch between sessions (#3490)
- [UI / Bug fix] Deleting a pinned session now removes it from the sidebar's Pinned section immediately, and sidebar rows no longer grow or shift while being deleted or renamed (#3492)
- [Chore] Faster initial load — syntax-highlighting language grammars now load on demand instead of all upfront. (#3496)
- [Bug fix / Feature] Antigravity (`agy`) sub-agents now reliably receive their first turn, get their approvals dismissed in the terminal, and stop showing as still-running after they have finished. (#3499)
- [Docs / Chore] qwen-native's terminal-start error label now reads "Qwen Code" (consistent with the other native harnesses) instead of "qwen". (#3500)
- [UI / Bug fix] Terminal view scrolling now works with macOS trackpads and with TUIs that enable mouse tracking at startup (OpenCode, Claude Code) (#3510)
- [Bug fix] Runners now retry login-page redirects with refreshed credentials instead of exiting, so a hosted session survives an expired bearer (e.g. after the machine slept through a token's lifetime) and reconnects on its own. (#3511)
- [UI / Bug fix] The chat view shows the "Starting up…" spinner while a message is waking a disconnected runner, instead of nothing until the runner boots; the sidebar row shows a spinner while a session is starting up (#3514)
- [UI / Bug fix] Renaming a session in the sidebar no longer hits the wrong row when the list reorders — row order holds while the pointer is over the list or a rename is in progress (#3515)
- [UI] Collapsed tool runs in the chat view are labeled by what they did ("Ran 1 shell command, read 2 files") instead of "See N steps" (#3518)
- [Feature] Sandbox dotfile hiding is now top-level only by default (opt into the full-tree walk with `cwd_hidden_scan_recursive: true`), and `mask_paths` hides named files or folders. Untrusted-tree sandboxes that relied on recursive masking should set `cwd_hidden_scan_recursive: true` on upgrade. (#3519)
- [UI / Bug fix] Cloning a session that runs in a git worktree now pre-fills the original repo with the worktree branch (instead of the worktree path as the working directory), and the clone dialog blocks creation when the picked working directory doesn't exist on the host (#3521)
- [Bug fix] `omnigent sandbox create --provider openshell` no longer crashes against openshell SDK >=0.0.86; workspace is configurable via `sandbox.openshell.workspace` or `$OMNIGENT_OPENSHELL_WORKSPACE` (defaults to `"default"`). (#3524)
- [Bug fix] Shared-session agents distinguish speakers without treating claimed roles as authorization. (#3527)
- [Bug fix] Direct `omnigent.llms.Client` Anthropic reasoning requests now select adaptive or fixed-budget thinking from live model capabilities. (#3529)
- [Bug fix / Docs / Chore] Model context limits now follow live provider catalog maximum-input metadata instead of adding output capacity or relying on stale built-in model IDs. (#3551)
- [Test/CI] N/A — internal CI enforcement only. (#3552)
- [Bug fix] User messages no longer risk a React hook-order crash when changing from a system marker to regular content. (#3553)
- [Bug fix / Docs / Chore] Pi now routes Databricks models through the API advertised by the live model catalog instead of a release-specific model allowlist. (#3572)
- [Bug fix] Bulk conversation actions now report structured errors when only some items fail. (#3573)
- [Feature] `omnigent import --force` replaces a previously imported chat with the latest local transcript. (#3576)
- [UI / Feature] Subagent graph panel now has zoom in, zoom out, and fit-to-view buttons for easier navigation of large agent trees (#3583)
- [UI / Bug fix] Change a session's model and effort from the config gear while the session is asleep — the change is saved immediately and applies when the next message wakes it. (#3584)
- [Feature / Test/CI] Add a Locust WebSocket load test (`dev/loadtest/`) with a one-command runner and result summary (#3591)
- [Feature] Sandbox now hides dotfiles (`.env`, `.aws`, `.ssh`, ...) under `write_paths` roots, not just `cwd` and `read_paths` (#3596)
- [Test/CI] N/A — internal CI enforcement only. (#3600)
- [Feature] GitHub policy blocks destructive operations (deletes) by default across MCP tools, `git push --delete`, and `gh * delete`; opt in with `allow_destructive: true`. (#3622)
- [Feature / Chore] The Cursor model picker now follows the models advertised by your installed Cursor CLI. (#3624)
- [Feature / Docs / Chore] Pi gateway sessions now list the models currently available from the workspace instead of a release-specific bundled menu. (#3629)
- [Docs / Chore] Setup and tool help no longer recommend release-specific model ids. (#3630)
- [Feature / Docs / Chore] Static model catalog responses now identify fallback ownership and advertise the current GPT 5.6 Sol, Luna, Terra, and GPT 5.5 Codex aliases. (#3632)
- [Feature] Catalog-backed model defaults now reuse a validated last-known-good provider catalog during transient upstream outages or empty responses. (#3641)
- [UI / Feature] Redesigned the sidebar's bulk-select bar and added per-section selection: pick sessions from the flat list or from within project folders (#3677)
- [Chore] Malformed tools, retry, and MCP YAML now fails with actionable parser errors instead of leaking untyped values. (#3705)
- [Chore] N/A — internal type-safety cleanup with no user-facing behavior change. (#3706)
- [Chore] N/A — internal ASGI type cleanup with no user-facing behavior change. (#3707)
- [Chore] N/A — internal migration decoding hardening with no supported-input behavior change. (#3708)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3729)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3734)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3735)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3736)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3737)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3739)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3740)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3741)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3742)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3743)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3744)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3745)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3746)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3747)
- [Bug fix / Chore] Managed Islo hosts now receive configured provider gateways during startup. (#3748)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3751)
- [Chore] N/A — internal type cleanup with no user-facing behavior change. (#3752)
- [Bug fix] Antigravity sessions without an explicit model now follow the installed SDK's current default. (#3762)
- [UI / Bug fix] Moving a session into a project updates the sidebar instantly instead of after a multi-second wait (#3784)
- [UI / Feature] `omnigent --help` now groups launch commands under a **Harnesses** section, colorizes the output, hides harnesses whose optional extra isn't installed (with a notice pointing at `omnigent setup`), drops the duplicate `update` alias line, and tidies the command descriptions (#3795)
- [Feature] `omni upgrade` now preserves requested extras for `uv tool` and `pipx` installs, supports `--extra`, `--target-version`, and `--dry-run`, and refuses to auto-upgrade `pip` / `uv pip` installs because those installers don't record extras. (#3796)
- [UI / Chore] `omnigent --help` command descriptions are tidied (harness rows read "Launch <Name> with Omnigent"), and the duplicate `update` alias line is hidden from the listing (#3797)
- [Bug fix] Prevent `list_files` from running without a conversation scope. (#3804)
- [Bug fix] Reject stale harness continuation requests that cannot resolve an agent model. (#3806)
- [Chore] N/A (internal type cleanup) (#3830)
- [Chore] N/A (internal type cleanup) (#3831)
- [Chore] N/A (internal type cleanup) (#3832)
- [Chore] N/A (internal type cleanup) (#3833)
- [Chore] N/A (internal type cleanup) (#3835)
- [Chore] N/A (internal type cleanup) (#3836)
- [Chore] N/A (internal type cleanup) (#3837)
- [Chore] N/A (internal type cleanup) (#3838)
- [UI / Bug fix] New sessions created inside a project appear under that project immediately instead of briefly showing under "Sessions" (#3869)
- [Bug fix / Breaking] The "GitHub Repo & Branch Access" and "Block Working Directory & Worktree (#3888)
- [Feature] claude-native sessions now derive their Working/idle status from Claude Code's own session file for faster, more accurate turn-edge detection (falls back to the terminal watcher on older Claude versions) (#3906)
- [UI] Tightened the sidebar's spacing so nav, sections, and session rows sit on a consistent vertical rhythm (#3908)
- [UI / Bug fix] Sidebar rows no longer stay highlighted in "Select sessions" mode unless explicitly selected (#3912)
- [UI / Feature] Opening a shell on a sleeping session now wakes its runner automatically instead of failing with "no runner available" (#3919)
- [Bug fix] Pi-native sessions now retain Omnigent system tools and the comment relay. (#3920)
- [Chore] N/A — no user-facing behavior change. (#3923)
- [Bug fix] Fixed a leak where native Codex sub-agents left orphaned `codex app-server` processes after idle reaping, TUI exit, runner shutdown, or a hard host/runner death (#3925)
- [Bug fix / Chore] OpenCode-native model options now fall back to the authenticated server catalog when CLI discovery fails. (#3926)
- [UI / Bug fix] The sidebar's My sessions / Shared with me switch stays visible while bulk-selecting sessions (#3927)
- [Feature] `omnigent diagnose` prints a secret-free environment snapshot (CLI/server versions, OS, auth mode) for bug reports (#3928)
- [UI / Bug fix] Hide the empty Projects header menu (⋯) when you have no projects (#3930)
- [UI] Moved the Chat/Terminal switcher for terminal-first sessions from the composer into the session header (#3931)
- [UI / Bug fix / Feature] Recent conversations reopen instantly while catching up without reordering live response output. (#3932)
- [Bug fix] `pip install` / `uv sync` no longer hangs when pnpm is provided by a corepack (#3986)
## [v0.7.0] — 2026-07-27
- [Bug fix] Hermes thinking now appears in mirrored web conversations. (#1645)
- [Bug fix] Image and file attachments now survive session relaunches on remote host runners; attachments that fail to load show a visible marker instead of silently disappearing. (#2085)
- [UI / Feature] Voice dictation in the composer now works in Electron, Firefox, and Chromium via optional server-side transcription (`omnigent[dictation]`) — local models, live streaming partials, audio never leaves your server. (#2093)
- [UI / Bug fix / Test/CI] Hide Claude task completion control messages from conversation history while preserving them for resume context. (#2104)
- [Bug fix / Feature / Docs] Operators can mount pre-created PersistentVolumeClaims (NFS/SMB/SAN) into Kubernetes sandbox runners via `sandbox.kubernetes.pvc_mounts` (read-only by default) (#2435)
- [Bug fix / Test/CI] `/compact` no longer races when multiple compact requests hit the same session at once (#2585)
- [Bug fix / Test/CI] `sys_call_async` / `sys_cancel_async` now consistently use `handle_id` as the cancel identifier. (#2586)
- [Bug fix / Test/CI] Runner idle timeout no longer kills sessions waiting on async tools, timers, or approval prompts (#2588)
- [Bug fix] Misconfigured runner tool policies deny tool calls instead of silently allowing them (#2589)
- [UI / Feature] Slash-command menus now match any part of a command's name, so `/using-superpowers` finds `/superpowers:using-superpowers` (#2655)
- [Bug fix] Host-launched runners now reuse delegated credentials instead of repeating Databricks authentication during startup. (#2762)
- [Feature] Projects are now a first-class entity with a `/v1/projects` CRUD API (create, list, rename, delete) and per-session membership. (#2765)
- [Feature] `omni usage` reports your LLM cost for today / the last 7 / 30 days, with a per-session per-model cost breakdown (#2787)
- [Feature / Chore] Native Claude sessions start faster by coalescing runner initialization into one handshake. (#2793)
- [UI / Bug fix / Feature / Docs / Test/CI] Claude-native launch and in-session pickers now share the selected host's live model catalog, including Claude Code's managed routes. (#2831)
- [Bug fix] Managed BoxLite sandboxes remain available after provisioning so agent launches can execute commands reliably. (#2846)
- [Feature] Server-side smart routing can now call an external `routes:select` router via `routing.provider: external`, with provider-agnostic auth (`api_key`) and model-name mapping (`model_prefix`) (#2864)
- [UI] Chat code blocks no longer load the syntax-highlighter engine until the first (#2886)
- [UI / Bug fix] The main chat "Working…" indicator now clears reliably when the session goes idle, instead of occasionally staying lit after a reply completes. (#2900)
- [Bug fix] The performance benchmark harness now records HTTP failures and continues the rest of the suite instead of aborting, and excludes fully-failed runs from the summary averages. (#2917)
- [Feature / Test/CI] Scheduled tasks can now be created without a workspace or a pinned host for non-code work (research, summaries, chat-only, MCP-only); an unset host runs on your live host at fire time, and an unset workspace defaults to the host's home directory. A pinned host is now authorized (existence + ownership) at create time rather than only at fire time. (#2946)
- [Feature] Set `OMNIGENT_CONTAINER_RUNTIME=podman` to use Podman (or another supported runtime) globally instead of Docker, without editing every agent's YAML. (#2949)
- [Bug fix] Sending a message to a session whose Claude Code terminal crashed no longer (#2951)
- [Bug fix] The desktop app now always quits within a few seconds even if its background cleanup stalls or the OS re-quit is dropped. (#2972)
- [UI / Bug fix] Messages send immediately when a session's only remaining work is a background job, instead of being held in the queue until it finishes (#2974)
- [UI / Feature] Desktop update notifications now appear in a native corner toast that works (#2975)
- [Bug fix / Chore] Runner startup no longer waits several seconds for Git's optional untracked-file cache probe. (#2976)
- [UI / Bug fix] Pi sessions now show reasoning while it streams and after conversation history reloads. (#2979)
- [Feature] The runner log now records why the runner exited (crash traceback, signal, idle timeout, tunnel close, or parent death) (#2985)
- [UI / Feature] Set up a missing agent from the New Chat dialog with a guided, step-by-step checklist (#2987)
- [UI / Bug fix / Feature] HTTP headers can now be set and edited for HTTP MCP servers in the session agent info panel. (#2989)
- Capped unbounded DB list queries in the permission store and reduced session opens in `check_access`/`get_permission_level` from 2–3 to 1. (#2995)
- Deleting a conversation with many descendants now issues a single FTS DELETE instead of one per descendant. (#2999)
- [UI / Bug fix] Android auto theme and system-bar icons now stay readable with both device themes and explicit in-app theme overrides. (#3006)
- [UI / Feature] 3D model files (STL, 3MF, OBJ) now render an interactive preview in the file browser (#3007)
- [UI / Bug fix] Subagents panel Graph View now shows the same status dot colors as List View (#3009)
- [Feature / Test/CI] Scheduled-task runs now transition to a terminal state (`succeeded`/`failed`) as soon as the dispatched turn finishes, instead of staying `running` forever; run history is readable at `GET /v1/scheduled-tasks/{id}/runs`. (#3014)
- [Feature / Chore] When enabled, new sessions receive concise semantic titles in the background without adding work or latency to the active agent turn. (#3024)
- [Feature] Offload dictation speech-to-text to a remote worker with (#3025)
- [Bug fix / Docs / Test/CI] Codex-native subagents now appear in the Agents panel with their live conversations. (#3028)
- [Bug fix] Credential proxy no longer attaches injected credentials to TRACE/OPTIONS requests, and the egress proxy now honors Max-Forwards as a conformant intermediary. (#3029)
- [Feature] Import local Qwen, Kiro, Pi, and Kimi coding chats into Omnigent (#3032)
- [UI / Feature] Press ⌘⌥V (Ctrl+Alt+V) to toggle voice dictation from anywhere; while dictating, Enter keeps the text and Esc discards it (#3044)
- [UI / Feature] Added: "Auto · smart routing" harness option in the new-chat picker — lets the intelligent router pick both harness and model based on the task description (#3045)
- [Feature] Import existing OpenCode chats, including files and tool activity, with `omnigent import` (#3046)
- [Bug fix] Dictation streams now reliably release their worker slot when a browser disconnects abruptly. (#3048)
- [UI] New-session composer moves harness configuration into a gear-icon modal, with a cleaner agent picker (needs-setup and custom agents folded into flyouts) and Smart Routing offered as a model option. (#3050)
- [Bug fix / Feature] The Slack bot can now run against an Omnigent server deployed on Databricks Apps, (#3051)
- [Feature] Sessions can now be filed into first-class projects via `PATCH /v1/sessions/{id}` and listed with `GET /v1/sessions?project=<name>`, which dual-reads first-class membership and legacy project labels. (#3053)
- [Bug fix / Test/CI] Fixed SDK session telemetry always recording `harness: null` in server deployments not started via the CLI. (#3054)
- [Bug fix] Databricks OAuth CLI profiles no longer fail with a misleading "malformed profile" error; the message now explains the real fix (install `omnigent[databricks]` or refresh the OAuth session). (#3059)
- [Bug fix] An idle runner shutting down after inactivity no longer shows a scary "disconnected" error — just send a message to wake it back up. (#3060)
- [UI / Feature] The sidebar now uses first-class projects: create empty projects, rename and delete them, and file sessions into them — while existing label-based projects keep working. (#3061)
- [UI / Bug fix] The "Host is offline — click to reconnect" prompt now appears in the composer's host badge instead of a separate banner below the composer (#3062)
- [Test/CI] N/A (test-only change) (#3063)
- [Feature / Docs / Test/CI] New `databricks_cli` credential-proxy type lets sandboxed agents use the Databricks CLI without the real token entering the sandbox (#3080)
- [UI / Feature / Test/CI] Polly sessions running on Claude SDK can start Goal mode from the chat composer. (#3084)
- [UI / Bug fix] The Configure agent modal's footer no longer shows a gray background band behind Cancel/Save (#3089)
- [UI / Feature] The Sidebar is more compact and polished, with clearer status indicators and richer session details on hover. (#3092)
- [UI] Reordered the project-folder header buttons (new-session before the menu), (#3096)
- [Chore / Breaking] `omni server start` is removed; use `omni server --background` to launch the (#3105)
- [Bug fix] `omnigent server --host 0.0.0.0` now enables accounts (login) mode automatically instead of silently 401-ing every request (#3107)
- [Feature] Projects can store default session settings (host, workspace, harness, model, …) via a new `config` field on the projects API. (#3108)
- [Bug fix] Databricks-served Claude models no longer break non-streaming responses (prompt-policy and smart routing) when returning typed content blocks (#3109)
- [Feature] `omnidev omnigent <args…>` runs an omnigent command against the current checkout's pod via `uv run omnigent`, with the pod's isolated env applied (#3110)
- [UI / Feature] Configure a session's model, effort, and smart routing mid-chat from a new gear icon in the composer (#3111)
- [UI / Feature / Test/CI] Add the `/tasks` Scheduled Tasks page with sidebar navigation, task rows, empty states, suggestion chips, create-dialog entry points, and Playwright E2E coverage. (#3112)
- [Bug fix] Reading image files in a Claude Code native session no longer bloats conversation history and breaks resume/compaction on large sessions (#3113)
- [Bug fix / Test/CI] The iOS app no longer follows cross-origin redirects when probing a newly approved server for the Databricks workspace mount, so a consented host can't redirect the probe to a different origin. (#3115)
- [Bug fix] Forked native sessions (Claude Code, Codex, Pi, Qwen) again resume with their prior conversation history. (#3116)
- [UI / Feature / Test/CI] Workspace pane icons now explain themselves on hover, with a cleaner right-side session layout and compact Share action. (#3122)
- [UI / Bug fix / Feature] Add a dialog for creating recurring scheduled agent tasks. (#3123)
- [UI] Sidebar session hover flyouts and rows now align with the project rows — matching flyout style, title size, and right-edge padding. (#3124)
- [Bug fix] Resuming a session with large images stored in history no longer overflows the context window or breaks compaction, on both the SDK and native Claude Code paths (#3133)
- [Feature] `omnigent session import` loads a `session export` JSONL back into a server as a new session (#3141)
- [Feature] Telemetry now records the agent name for Polly and Debby sessions. (#3152)
- [Chore / Breaking] `omni integration slack start` is removed; use `omni integration slack --background` to launch the (#3153)
- [UI / Bug fix / Docs / Chore] Slack device login now requires a fresh password at the consent screen, closing a device-code phishing gap where an already-signed-in user could approve a login by reflex. (#3156)
- [Feature] `omnigent claude` keeps tool search enabled when launched with `CLAUDE_CODE_USE_GATEWAY=1`. (#3161)
- [Test/CI] Fix `test_session_stream_emits_heartbeat_on_idle` after `_session_labels_for_runner_spawn` was extracted into `omnigent.runner.native.orchestration`; patch the heartbeat cadence on `omnigent.runner.app` where it is defined and consumed. (#3163)
- [Bug fix] Crash reports are no longer lost when a process crashes more than once in the same second. (#3173)
- [Bug fix] Custom codex-native agents launch on the model declared in the agent spec (`executor.model`) instead of silently falling back to the provider default (#3175)
- [Bug fix] Single-file agent YAMLs that nest the executor under `type:`/`config:` (the bundle config.yaml shape) now fail at load time with the correct flat spelling, instead of silently running a harness inferred from the model prefix (#3178)
- [UI / Bug fix / Feature] Use native Codex goal mode from Polly's Goal control (#3181)
- [UI / Bug fix] Renaming a session now updates the name in the sidebar instantly instead of after a short delay. (#3185)
- [UI / Bug fix / Feature / Test/CI] Edit scheduled tasks and type exact run times, with a scrollable time picker and a consistent, fully-visible dialog. (#3186)
- [UI / Feature] Pinned sessions now persist server-side per user, so pins follow you across devices and browsers. (#3189)
- [Feature] New sessions now receive concise semantic titles automatically without additional configuration. (#3191)
- [Test/CI] `/rerun` PR comment re-runs failed CI on the current commit without dismissing approvals (#3195)
- [Feature] Native Codex sessions can now receive concise automatic background titles. (#3199)
- [Bug fix] Qwen3, inkling, and other non-OpenAI models now work in the Pi SDK executor harness (#3203)
- [Chore / Breaking] Slack-on-Databricks deploy: renamed `OMNIGENT_SLACK_WEBAUTH_BASE_URL` to `OMNIGENT_SLACK_DATABRICKS_APP_URL` (`--app-url`), removed the `WEBAUTH_PORT` / `DATABRICKS_WORKSPACE_HOST` overrides, and dropped deploy-time `uv lock` in favor of in-container `uv run`. (#3206)
- [UI / Bug fix] Sidebar session titles use the available space cleanly and reveal branch and action details only when needed. (#3208)
- [UI / Bug fix] Removed the redundant "Create new project" option from the sidebar project picker — create projects with the + icon next to Projects (#3210)
- [Feature] Smart routing now activates automatically when a server `llm:` block or an external `routing:` block is configured — no `OMNIGENT_SMART_ROUTING` env var needed (#3215)
- [UI / Feature / Test/CI] Scheduled task rows now show when each task will next run ("Next run in 15h") and a "Run now" action in the ⋯ menu to fire a task immediately, with refreshed row styling. (#3218)
- [UI / Feature] Projects now carry default session settings (host, working directory, agent, optional git worktree) that pre-fill the new-session composer. (#3221)
- [Bug fix] Fixed per-model cost attribution for native harnesses so a session's per-model (#3223)
- [UI / Bug fix] Codex task plans now stay in Tasks instead of being duplicated in chat. (#3249)
- [UI / Bug fix] Smart Routing no longer appears in the model dropdown for native terminal sessions (Claude Code, Codex, Pi), where it had no effect (#3259)
- [Bug fix] Sandboxed agents can now run tools managed by update-alternatives (awk, python3, editor, and similar) on Linux. (#3263)
- [Bug fix] Egress proxy now trusts corporate/MDM CA roots installed under the system `capath` directory, so TLS to hosts behind a corporate MITM works from a sandboxed agent. (#3264)
- [Bug fix] Large historical attachments no longer inflate replay and compaction context as inline base64 text. (#3267)
- [Docs] Contributors can now use `omnidev` as the documented worktree-safe local testing flow. (#3277)
- [Bug fix] Custom OpenAI Agents can use Unity AI Gateway Model Services with fully qualified model names when a Databricks provider or profile is configured. (#3288)
- [UI / Bug fix / Feature] Configure recoverable dangerous shell commands to ask for approval or deny execution, while always blocking catastrophic operations. (#3297)
- [Bug fix / Feature] Pi harness now routes kimi, inkling, GLM, qwen3, Gemini 3+, and Llama through the correct AI Gateway endpoints, fixing "Stream ended without finish_reason" errors and ensuring `system.ai.*` ids are used throughout. (#3307)
- [Feature] Add a top-level `justfile` with recipes for launching the iOS Simulator, running the Android debug build, starting the omnigent dev pod, and running pre-commit/lockfile normalization. (#3310)
- [UI / Bug fix] Sidebar header action buttons are now vertically centered with section titles. Session row hover is smoother, and active items no longer flash when hovered. (#3311)
- [UI / Bug fix] Aligns project folder icons and color with the rest of the sidebar. (#3317)
- [Bug fix] Unsupported Claude Code slash commands are now escaped and sent as regular user messages instead of leaving the native terminal in an undriveable state. (#3319)
- [Feature] `omnigent server` now tells users they can set `OMNIGENT_AUTH_ENABLED=0` to override automatic multi-user mode when binding to a non-loopback interface. (#3320)
- [Chore / Breaking] [Breaking] The deprecated `OMNIGENT_ACCOUNTS_ENABLED` environment variable has been removed; use `OMNIGENT_AUTH_ENABLED` instead. (#3322)
- [UI / Bug fix] Fixed pinned sessions being lost when the web UI was updated before the server. (#3323)
- [UI / Feature] Automations list: tasks now render as cards and show a live-updating relative next-run time ("Next run in 3 hours"). (#3324)
- [UI] Settings → Appearance no longer has a separate Sidebar font size control, and a new "Reset to defaults" button restores every appearance preference after confirmation. (#3326)
- [UI] Starting a session inside a project now names the project in the new-session (#3327)
- [UI] The Files workspace tab now uses a stacked-files icon. (#3329)
- [UI / Feature] Automations: scheduled tasks can now pick a model and reasoning effort in the create/edit dialog (defaults to the agent's settings). (#3331)
- [UI / Bug fix] Fixed sessions pinned in the updated web UI being lost after the server was updated. (#3332)
- [UI / Feature] Native harness setup now checks the installed CLI version and prompts to upgrade if it is too old. Cursor's missing-binary case is normalized to the same structured `binary-missing` signal as the other CLI-backed native harnesses. (#3335)
- [Bug fix] Pi sessions now show a clear error when their Databricks login has expired, instead of silently accepting messages with no reply (#3336)
- [Test/CI] N/A (internal CI change). (#3338)
- [Bug fix] claude-sdk harness now surfaces harness-level failures (expired login, auth error) as structured errors instead of storing them as assistant messages. (#3342)
- [UI] Align the sidebar brand row with the rest of the navigation. (#3346)
- [UI / Bug fix] Short links like `#3090` in chat markdown tables no longer stack one character per line (#3350)
## [v0.6.0] — 2026-07-21
- [Bug fix] `sys_os_shell` commands no longer inherit omnigent's own `PYTHONPATH` entry, so project subprocesses resolve their own installed packages instead of omnigent's. (#1861)
- [Bug fix] Stopping a Goose turn no longer leaves stale ACP session state behind when the subprocess is terminated. (#1928)
- [Bug fix] Cancelling or deleting a session during agent cold start no longer leaks the harness subprocess. (#1982)
- [Bug fix] macOS sign-in with a hardware security key (e.g. YubiKey) works again. (#2036)
- [UI / Bug fix] Tool-call cards keep their live spinner when a superseded response terminates mid-turn (fixes a first-turn "no spinner" on native-terminal harnesses). (#2045)
- [Bug fix] hermes-native tool-call cards now show a live spinner and elapsed timer while a tool runs, including on the first turn. (#2046)
- [Bug fix] Fix blank white page on Safari/iPadOS older than 16.4 caused by regex lookbehinds on the web UI boot path (#2105)
- [UI / Feature / Chore] A project's "new session" pencil now prefills the composer from the project's latest session — host, repo, agent, and a fresh git worktree branch — so starting a new chat in a project is one prompt away. (#2133)
- [UI / Feature] Filter the Archived sessions view by project (#2134)
- [Bug fix / Docs / Test/CI] Polly now verifies pytest test-count discrepancies against collected cases before recording miscount/fabrication claims. (#2140)
- [Bug fix] Evict the cached claude-sdk client when a turn is cancelled so the session recovers on the next turn (#2169)
- [UI / Bug fix] Tapping an Android notification now opens the waiting session (or the inbox). (#2210)
- [Bug fix] Fixed the headless Hermes harness registering no Omnigent MCP server (agent had no builtin tools). (#2216)
- [Feature / Breaking] Ids are now bare 32-character hex (no `conv_`/`ag_`/`host_` prefixes) stored as 16-byte binary; existing prefixed ids in URLs, configs, and clients keep working. (#2228)
- [Bug fix] `sys_list_models` no longer misreports codex `cli-config` providers and cursor workers as having no usable model provider. (#2237)
- [Bug fix] Cursor sessions now run shell tools in the declared workspace directory instead of the runner's working directory (#2244)
- [UI / Bug fix / Feature] Conversation view gains a turn-rail minimap for jumping between messages (#2285)
- [Feature / Docs] `sandbox.host_config` in the server config injects verbatim host `config.yaml` content (e.g. a `providers:` gateway block) into managed sandboxes before the host starts (#2306)
- [Bug fix] `omnigent server` now accepts documented boolean server-config values like `sandbox.kubernetes.in_cluster: false` instead of rejecting them at startup. (#2314)
- [Bug fix] Chat sessions now recover automatically from a brief reverse-proxy 404 during a backend restart, instead of freezing until a manual page refresh. (#2316)
- [UI / Bug fix] Bulk select-all/delete/archive in the sidebar now correctly handles collapsed sections (#2377)
- [Bug fix / Feature] `omnigent run --harness <name>-native` now launches every registered native harness with matching prompt and model behavior. (#2379)
- [Feature / Docs / Test/CI] Harness bench now reports Omnigent MCP relay support separately from vendor-native tool calling. (#2380)
- [Bug fix] Sessions are now stopped server-side before archive or delete, so SDK and API callers get the same stop-first behavior the web UI provides. (#2400)
- [UI / Bug fix / Test/CI] Sidebar session tabs no longer overflow their background on narrow widths. (#2425)
- [Bug fix] Local server startup now works when a system proxy intercepts loopback traffic. (#2433)
- [Bug fix] Native TUI harnesses (opencode, pi, hermes) no longer render multibyte UTF-8 as mojibake on deployments whose environment lacks a UTF-8 `LANG`/`LC_ALL`. (#2440)
- [Bug fix] Claude native sessions now back off rejected cost updates instead of retrying on every poll. (#2453)
- [Feature] Server now collects opt-out usage telemetry for session created, stopped, and deleted events to help understand product usage patterns. (#2457)
- [UI / Bug fix] Confirming a Japanese IME conversion with Enter no longer submits the session rename or new-project name inputs (#2459)
- [Feature] Server `llm:` config accepts an optional `fallback_models:` list; LLM-backed policies now retry alternate models before failing closed. (#2462)
- [Feature] Harden the LLM prompt-classifier policy against prompt injection by spotlighting untrusted content behind an unguessable per-evaluation marker. (#2463)
- [Feature / Docs / Test/CI] Harness bench now reports whether forked sessions retain and replay their source conversation history. (#2472)
- [Bug fix / Test/CI] Harness benchmark live tables now distinguish not-applicable capabilities from skipped probes. (#2475)
- [UI / Bug fix] New sessions respond faster to their first message, and the "Working…" indicator now appears immediately on the first turn instead of staying dark while the runner finishes connecting (#2478)
- [Bug fix] A policy-denied message (e.g. hitting the cost budget) now shows immediately instead of only after a page refresh (#2481)
- [Bug fix / Feature / Docs / Test/CI] Codex sessions using Databricks custom model aliases can surface reasoning summaries, and the harness benchmark now reports observable reasoning support. (#2482)
- [Feature / Docs / Test/CI] The harness benchmark can run selected capability dimensions and pin an exact model directly in each harness argument. (#2485)
- [Bug fix / Test/CI] Polly Cursor workers no longer ask for tool permissions by default (YOLO / auto) (#2493)
- [Feature] Polly defaults to Sonnet 5 for its brain and Claude Code workers, and Cursor Grok 4.5 for Cursor workers (#2500)
- [Bug fix] Polly Cursor workers default to grok-4.5 (#2503)
- [Bug fix] Polly Claude brain and Claude Code workers default to the sonnet alias (#2504)
- [Bug fix] Polly Claude brain and Claude Code workers inherit the provider default model again (#2507)
- [Bug fix] Codex-harness sub-agents no longer crash on macOS when the runner inherits a read-only working directory (e.g. `/` from the desktop app). (#2512)
- [UI / Feature / Test/CI] New chats can start with the Workspace panel collapsed via Appearance settings (#2516)
- [Bug fix] `sys_advise_models` is only offered to agents when intelligent routing is configured on the server. (#2517)
- [UI / Feature / Test/CI] OpenCode-native sessions now show their available models in the web model picker. (#2519)
- [Bug fix] Pi's `/model` command now shows all LLM models available on your Databricks workspace (fetched live from the serving-endpoints API) instead of a hardcoded list (#2525)
- [UI / Feature] The changed-files panel now shows a per-file line-change count (`+N −M`) beside each file. (#2526)
- [Bug fix] Telemetry rollout percentages now honor 0% and 100% exactly. (#2528)
- [Bug fix] Pi no longer shows a blocking "Trust project folder?" dialog when launched via Omnigent in a project with `.pi/` settings or extensions (requires Pi 0.79+) (#2529)
- [Feature / Docs / Test/CI] Harness metadata can now declare resume, steering, live queue, image, and compaction support for capability conformance checks. (#2530)
- [UI] Admin Settings pages (Members, Policies, Sharing) now share the same left and top alignment as the rest of Settings. (#2532)
- [Bug fix] Pi's `/model` command now shows GLM, Llama, Qwen, and other non-GPT Databricks models alongside Claude and GPT models (#2534)
- [UI / Bug fix] The Members and Sharing settings pages and the Share buttons are hidden in single-user mode (#2536)
- [Bug fix] Pi's `/model` command now correctly lists all available Databricks models when using the AI Gateway (cli-config) setup — the NXDOMAIN error that caused single-model fallback is fixed (#2540)
- [UI / Feature] Native Pi sessions can now switch model mid-session from the web composer, with the picker scoped to your logged-in models and synced to in-terminal `/model` changes (#2543)
- [UI / Feature] Add an opt-in "Hide unconfigured harnesses" setting that filters the new-chat picker to harnesses set up on the selected host (#2544)
- [Bug fix] Sending a message to an idle native Pi session no longer gets stuck in the "queued" state until you switch tabs (#2545)
- [Feature / Docs / Test/CI] Add `omnigent uninstall` for safely removing the CLI, installer-managed shell/config edits, and optionally local Omnigent state. (#2550)
- [Bug fix] Pi's `/model` command now shows all available Databricks models for ucode / Codex app profile-switched setups (#2552)
- [UI / Feature] Added a Nord color theme (arctic frost-blue palette) to the Appearance settings palette picker. (#2561)
- [Bug fix] Fix `uv tool install` of the `databricks` extra failing to build pyarrow from source on Python 3.14 (#2563)
- [Bug fix] Sessions list no longer does a workspace-wide id pre-fetch for the archived filter, restoring the single indexed query on the sidebar path. (#2568)
- [Bug fix] The CLI now shows a clear error instead of a raw traceback when the server is too slow or unreachable while starting a session. (#2572)
- [Bug fix] Pi harness: Databricks GLM/DeepSeek reasoning models no longer fail with "Stream ended without finish_reason" (#2573)
- [UI / Bug fix] Clicking a pinned session that belongs to a project no longer re-opens the project section after you've collapsed it. (#2583)
- [Bug fix] Claude-native messages that fail to reach Claude (or stall unsent) now surface the failure instead of silently disappearing. (#2591)
- [UI / Bug fix] The file editor now follows Omnigent's selected web and desktop color theme instead of the operating-system theme. (#2594)
- [UI / Feature] Hovering a pinned session now shows its project name in a flyout (#2595)
- [UI] Sidebar rows, section headers, and nav items now line up on a consistent grid (#2596)
- [Bug fix] Queued follow-up messages in the Pi web UI now dispatch immediately when Pi finishes replying (#2597)
- [UI / Bug fix] Pinned-session project flyout no longer opens (and gets stuck over the chat) when tapping a session on mobile (#2599)
- [UI / Feature] Diff viewer gains a "Wrap lines" toggle, and Find / Download / diff toggles now live in a single "⋯" View settings menu (#2600)
- [UI / Bug fix] Mobile session menu opens the project picker in place instead of an off-screen side flyout (#2602)
- [Bug fix] `sys_session_send` now reliably accepts model overrides when spawning named sub-agent sessions with a specific model (#2603)
- [Chore / Breaking] `omnigent[memory]` is renamed to `omnigent[hindsight]`; the Hindsight memory extra is now the canonical name. (#2605)
- [Bug fix] `omnigent host status` now hides the session table by default (use `--sessions` to show it), making the command significantly faster. (#2606)
- [Feature] `omnigent://<hostname>/c/<session_id>` links open that session in the desktop app, reusing an open window on that server in-place (#2607)
- [Bug fix] Fix native Claude sessions overflowing the context limit on reconnect when the conversation contained screenshots (#2609)
- [Feature] Session creation requests from the web UI now send an explicit `X-Omnigent-Client` header, enabling the server to record the exact client surface (web/desktop/ios/android) in telemetry without relying on User-Agent parsing. (#2615)
- [UI / Bug fix] Mobile: the Chat/Terminal toggle no longer reappears over the sidebar when opening a session's kebab menu (#2617)
- [UI] Project sidebar headers show a chevron in place of the folder icon on hover instead of a persistent caret next to the name (#2618)
- [UI / Feature] PDF files now preview inline in the file viewer, with scrollable pages and zoom (#2619)
- [UI / Bug fix] Find in file now toggles the editor's find bar closed on a second click, and its close button no longer shows a stray keyboard-hint tooltip (#2621)
- [UI / Bug fix / Feature] Find in file now works in the markdown editor — highlights and cycles matches, including terms split across bold/italic formatting (#2628)
- [Bug fix] Clicking an Omnigent session URL printed inside the embedded terminal now stays in the web app instead of opening a duplicate browser tab/window. (#2639)
- [Docs] Clarify optional install extras in the quick-start docs. (#2640)
- [Bug fix] `use_responses: false` now correctly selects the Chat Completions API in standard `config.yaml` bundles. (#2641)
- [Bug fix / Feature] Import existing Claude Code or Codex chats with `omni import`. (#2649)
- [UI / Feature] Customize a built-in color theme with shared light/dark accent, tint, contrast, and two-rail sidebar translucency controls. (#2650)
- [UI / Bug fix] Pinned sessions no longer show duplicate rows after upgrading — legacy pin ids are migrated to the new id format automatically. (#2651)
- [UI / Feature] Long-running Codex commands now stream their output into the web tool card while they run. (#2652)
- [UI / Feature] Randomize accent and background tint colors with a playful new dice button. (#2653)
- [Feature] `omnigent://<hostname>/c/<session_id>` links open that session in the iOS app, reusing the open window on that server in-place (#2661)
- [Bug fix] Pi's `/model` command now shows and correctly routes GPT, Kimi, Llama, GLM, and Gemini 3.x models alongside Claude (#2665)
- [UI / Bug fix] The Add Global Policy dialog now shows all available policies, including ones already applied. (#2668)
- [UI / Bug fix] The Add Policy dialog for sessions now shows all available policies, including ones already applied. (#2670)
- [Bug fix] Deleting or archiving a session now also stops any still-running sub-agent children, instead of leaving them running as unreachable orphans. (#2673)
- [UI / Feature] Find in file now works in the markdown and notebook preview — highlights and cycles matches in the rendered document (#2674)
- [Bug fix] Codex sessions through a Databricks gateway profile no longer fail with "Invalid Token" when the environment's `DATABRICKS_HOST` points at a different workspace than the profile (#2675)
- [UI / Feature] The file panel (changed files, browsing, diffs, and file contents) stays viewable when a session's runner is offline but its host is still connected — no need to send a message to wake it. (#2676)
- [UI / Feature] Select text in a PDF preview to add review comments with inline highlight overlays. (#2677)
- [UI / Feature] Codex plans now show up in the Tasks panel, the same way Claude Code todos do (#2678)
- [UI / Bug fix] Token expiration and other harness startup errors now appear immediately in the chat transcript instead of requiring a page reload. (#2681)
- [UI / Feature] Switching back to a recently-opened chat now renders instantly from cache instead of blanking and reloading over the network. (#2688)
- [UI / Bug fix] Fixed the comment box being hidden behind the keyboard when commenting on a file in the iOS app (#2694)
- [Bug fix] Host-bound sessions relaunch immediately after a Stop or host restart instead of waiting out the runner-connect grace. (#2699)
- [Test/CI] Add a `checkout_sha` input to the Benchmark workflow so nightly benchmarks can be run ad-hoc against a specific commit. (#2700)
- [Bug fix / Test/CI] Fresh installs no longer resolve an OpenAI SDK combination that crashes every OpenAI Agents harness turn. (#2713)
- [Feature] `omni import --harness <claude|codex> --last N` can now import up to 50 recent local chats at once. (#2724)
- [UI / Bug fix] Cursor setup now distinguishes CLI readiness from SDK API-key configuration and shows install/login guidance before web chat launch. (#2733)
- [UI] The Gateway option in model setup no longer cites OpenRouter as an example — use the dedicated OpenRouter option instead (#2738)
- [Bug fix] Sandboxed (`darwin_seatbelt`) agents now boot when the helper interpreter or the wrapped CLI is reached through symlinks — uv-installed Pythons and the standalone Claude CLI no longer die with `Operation not permitted`. (#2743)
- [Bug fix] A sandboxed claude-sdk agent no longer dies at connect time when its sandbox can't wrap the Claude CLI — it starts with native tools disabled (file/shell access stays sandboxed via the `sys_os_*` tools) and logs why. (#2749)
- [Feature] `omnigent setup` can now install Hermes directly before configuring its model provider. (#2751)
- [Bug fix] Sandboxed (`darwin_seatbelt`) agents no longer die at startup with "No usable temporary directory" — the private scratch tmpdir is now granted in the sandbox profile before the process is jailed. (#2759)
- [Feature / Docs / Test/CI] Benchmark reports now measure cold restarts of existing sessions whose runner is offline. (#2761)
- [Docs / Chore / Test/CI] Website release posts now use the narrative, prose-driven MLflow style (#2764)
- [Bug fix] opencode-native carries the user config's model default into the synthesized config when no model is pinned (#2775)
- [Feature] New sessions automatically get concise, useful titles generated by the agent already handling the conversation. (#2778)
- [Bug fix] Forking a Claude Code session onto pi (or other CLIs) no longer fails to start with `required_terminal_exited` (#2780)
- [Bug fix] Native Codex and Claude now resolve their CLI from common global install dirs and an `OMNIGENT_CODEX_PATH` / `OMNIGENT_CLAUDE_PATH` override when it isn't on the host daemon's PATH (#2788)
- [Feature] Nightly benchmark report now includes server CPU% and RSS memory usage (mean/min/max) alongside latency metrics. (#2795)
- [Bug fix] The changed-files panel now shows per-file line-change counts even when the list is served by the host (runner offline). (#2802)
- [Bug fix] Harness readiness now finds CLIs installed in common global dirs (nvm/npm/homebrew) that aren't on the host daemon's PATH, matching what the launch resolves (#2805)
- [UI / Bug fix / Feature] Config-file policies (declared via `omni server -c`) now appear in the admin Global Policies page as read-only entries labeled "Config". (#2807)
- [Bug fix] Codex sessions now inherit global `AGENTS.md` instructions. (#2809)
- [UI / Feature] Switching back to a recently-opened chat now renders instantly from cache instead of blanking and reloading over the network. (#2810)
- [Bug fix] Allow the ACP prompt deadline to be configured without changing its default. (#2817)
- [Feature] The Slack bot now handles tool approvals and questions inline — Approve/Deny (#2820)
- [UI / Feature] The Share dialog now shows a QR code so you can open a session in the mobile app by scanning it with your phone. (#2824)
- [Bug fix] Harness setup warnings now clear automatically without reconnecting the host. (#2828)
- [UI / Feature] Android app shows a floating server switcher pill with a dropdown menu for quick server switching (#2829)
- [Feature] Crash reports now show a calm, branded screen with a compact traceback and a one-tap prompt to file a pre-filled GitHub issue (#2841)
- [UI / Bug fix] Parallel sub-agent dispatch now explains when to use distinct task titles instead of accidentally reusing and serializing one child session. (#2860)
- [Bug fix] Fixed a context-window overflow on a live-streamed turn, leaking its harness process instead of ending the turn cleanly. (#2869)
- [Bug fix] Clicking "Stop session" no longer shows a spurious "runner disconnected" error (#2903)
- [UI / Bug fix] New-session Send button shows a spinner while the session is being created, instead of looking frozen (#2907)
- [Feature / Breaking] Per-harness startup command/args overrides via a polymorphic `harness:` config key + `OMNIGENT_<NAME>_PATH` env-var standardization (legacy `HARNESS_*_PATH` deprecated, slated for v0.8.0 removal) (#2933)
- [Bug fix] `omnigent[memory]` is restored as a backwards-compat alias for `omnigent[hindsight]` (the `memory` extra was dropped by #2605); it will be removed in 0.70. (#2938)
- [Feature] `OMNIGENT_SESSION_RENAME=on` re-enables the automatic first-turn session rename (disabled by default) (#2944)
## [v0.5.1] — 2026-07-10
- [UI] The desktop Browser tab is now hidden on older desktop builds that don't support the embedded browser, instead of showing a tab that does nothing (#2393)
| `in_cluster` | Optional cluster-config source: `true` (in-cluster SA only), `false` (kubeconfig only), omit (try in-cluster, then kubeconfig). |
| `kubeconfig` | Optional kubeconfig path for the out-of-cluster fallback (env: `OMNIGENT_KUBERNETES_KUBECONFIG`). |
| `pvc_mounts` | Optional pre-created PersistentVolumeClaims mounted into every runner Pod — see [Persistent storage mounts](#persistent-storage-mounts-pvc_mounts). |
## Persistent storage mounts (`pvc_mounts`)
Runner Pods are ephemeral by design — the workspace lives on an `emptyDir` and
dies with the Pod. To expose durable data (datasets, model caches, shared
output directories) mount pre-created PersistentVolumeClaims:
1. Create the PV/PVC **in the runner namespace** (`omnigent-sandboxes`) out of
band — via your GitOps repo, with whatever backend your cluster provides
(NFS/SMB CSI drivers, SAN, cloud disks). Omnigent only references the claim;
it never creates volumes, so the server RBAC stays unchanged.
2. List the claims under `sandbox.kubernetes.pvc_mounts` (see
`sandbox-config.yaml`). Mount paths may not overlap `/home/omnigent`, the
OS directories, or their ancestors (e.g. `/home`, `/var`) — the server
rejects such config at startup.
Caveats:
- **Multiple runners share writable claims concurrently** — use a
`ReadWriteMany`-capable backend (NFS/SMB/CephFS) for anything writable, and
prefer `read_only: true` (the default) everywhere else: a writable shared
mount lets one session's agent read and modify what another session wrote,
and anything written there outlives the Pod and its launch token.
- Runner Pods run as uid/gid 1000660000 with `fsGroup`. NFS `root_squash` and
SMB ownership mapping must permit that identity (export to the uid, or use
CSI mount options like `uid=`/`gid=` for SMB); `fsGroupChangePolicy:
OnRootMismatch` avoids re-chowning large exports on every start.
-`ReadWriteOnce` claims pin all runners to one node — combine with
`node_selector` deliberately, or the second Pod sits `Pending`.
- A mount visible in the Pod is not automatically visible to a harness's own
OS-level sandbox (OmniBox path grants are separate).
To verify `host_config` end to end against a live cluster, run
3. Slack server shows the verification link in the setup modal (initiator
only). The device_code is NOT included — it never leaves the server
pair; only the user_code (in verification_uri_complete) does.
3. Slack server shows the verification link (verification_uri_complete,
code prefilled for one-click) in the setup modal (initiator only),
plus the user_code so the user can confirm the match. The
device_code is NOT included — it never leaves the server pair.
4. User clicks → Omnigent consent page (verification_uri).
Browser authenticates via the server's accounts provider.
Page shows: "<client_id> is requesting permission to act as YOU
(alice@example.com) on this Omnigent server. [Approve] [Deny]"
plus a warning to approve only a login the user personally started.
The page REQUIRES a login started for THIS flow: if the browser's
session predates the grant (session iat < grant.created_at), it
bounces through the login page with ?reauth=1 — which forces a fresh
password entry even for an already-signed-in user — and returns here.
Once re-authenticated, the page shows: "<client_id> is requesting
permission to act as YOU (alice@example.com) on this Omnigent server.
[Approve] [Deny]" plus a warning to approve only a self-started login.
The forced re-auth means a grant can't be approved by one reflexive
click on a link the user didn't personally start (see threat #2).
5. User approves → the grant is bound to the authenticated identity
(alice@…). client_id is recorded for display/audit only, never as
@@ -197,9 +189,8 @@ cli-ticket flow and never mounts these routes; header mode has no
server-mintable identity — see `create_device_auth_router`, which raises if
constructed for any other source). The `device_grants` table is created
unconditionally by the migration regardless of the flag; only the router
mount is gated. This router also **owns**`mint_delegated_token` and
`DELEGATED_SCOPE` (moved here from `oidc.py`, which retains only
`mint_session_token` / `mint_session_cookie`).
mount is gated. This router **owns**`mint_delegated_token` and
`DELEGATED_SCOPE`.
-`POST /oauth/device/authorize` — **public** (rate-limited). Generates a
high-entropy `device_code` (`secrets.token_urlsafe`, stored **hashed**), a
@@ -275,9 +266,8 @@ HS256 shape (so `_check_cookie` accepts them) plus four delegated-only claims:
## Slack-side changes
- **`oauth.py` (new)** — device-authorize → post ephemeral link → poll token
endpoint (respecting `interval` / `slow_down`) → store tokens. Replaces the
`AuthRequiredError` dead-end.
- **`oauth.py`** — device-authorize → post ephemeral link → poll token
endpoint (respecting `interval` / `slow_down`) → store tokens.
- **`omnigent.py`** — attach `Authorization: Bearer` per
`(server_url, slack_user_id)`; on 401, refresh once and retry; on refresh
failure, surface a re-login prompt. `OmnigentClientPool` keys clients by
@@ -285,8 +275,8 @@ HS256 shape (so `_check_cookie` accepts them) plus four delegated-only claims:
- **`store.py`** — new `oauth_tokens` table `(team_id, user_id, server_url)` →
access/refresh **encrypted at rest** (key from env / secret manager, never in
the DB). `/omnigent logout` → `POST /oauth/revoke` + local delete.
- **`setup.py`** — validation uses the user's token; auth-enabled servers become
supported rather than rejected.
- **`setup.py`** — validation uses the user's token, so auth-enabled servers
are supported.
- **`config.py`** — holds the local encryption key for token storage.
## Security analysis
@@ -294,7 +284,7 @@ HS256 shape (so `_check_cookie` accepts them) plus four delegated-only claims:
| # | Threat | Mitigation |
|---|--------|-----------|
| 1 | `device_code` leak → token theft | Never transits Slack or the user — only `verification_uri_complete` (a `user_code`) does. Stored hashed; single-use. |
| 2 | Link misdelivery / phishing another user | Link shown to the initiator only (in their own setup modal). Consent page names the exact Omnigent identity the grant will act as and the requesting `client_id`, and warns to approve only a self-initiated login. |
| 2 | Link misdelivery / phishing another user | Link shown to the initiator only (in their own setup modal). **Consent requires a login started FOR this flow: the consent page rejects a session whose `iat` predates the grant and bounces through the login page with `reauth=1`, forcing a fresh password entry even for an already-signed-in user.** So an attacker-initiated flow can't be approved by a single reflexive click — the victim must deliberately re-enter their password against a screen naming the exact Omnigent identity and requesting `client_id`. The gate is enforced on both the consent GET and the approve POST. |
| 3 | Anyone can initiate/poll (public client) | Cheap `pending` state grants nothing until an authenticated user approves. `POST /oauth/device/authorize` is rate-limited per client IP (10/60s → 429 `slow_down`); short (10 min) `device_code` expiry; `slow_down` enforced server-side on aggressive polling; expired grants purged opportunistically. |
| 5 | Compromised Slack server acts as all users (inherent to delegation) | Reduced scope (no admin), short TTL + refresh rotation, per-grant revocation, **absolute grant lifetime (30 d) enforced on refresh** so even an un-revoked grant dies, and an `act`-claim audit trail. |
@@ -315,6 +305,17 @@ token.
When no client secret is configured the endpoints are **public**, so
initiation is open — the defense is layered, not a gate:
- **Forced re-authentication at consent.** Consent requires a login started for
THIS flow: the consent page (and the approve POST) reject a session whose
`iat` predates the grant's `created_at` and bounce through the login page with
`reauth=1`, which forces a fresh password entry even for an already-signed-in
user. This defeats the reflex-approve variant of the attack — a victim handed a
one-click link (even one with the code prefilled) still can't bind the grant
without deliberately re-entering their password against a screen naming the
exact identity and client. (`device_auth.py``_session_iat` + the
`reauth=1` bounce; `LoginPage.tsx` suppresses its already-signed-in
auto-return under `reauth=1`.) The prefilled one-click link is therefore
retained for convenience — the re-auth step, not code handling, is the gate.
- The consent page prominently **warns** the user to approve only a login they
personally started and to match the code shown by the application.
- The delegated scope excludes admin / user-management endpoints.
@@ -322,6 +323,11 @@ initiation is open — the defense is layered, not a gate:
grant self-expires even if never revoked.
- Initiation is rate-limited per IP; nothing is granted until a real user
authenticates and approves in their own browser.
- **Startup warning.** When the grant is mounted on a multi-user (accounts)
server with `OMNIGENT_DEVICE_CLIENT_SECRET` unset, the server logs a loud
warning at startup that the authorize endpoint is public — nudging the
operator to opt into the secret rather than leaving initiation open unknowingly
(`app.py`, at the device-router mount).
Setting `OMNIGENT_DEVICE_CLIENT_SECRET` closes initiation entirely to
unauthorized callers: without the matching `X-Omnigent-Client-Secret` header,
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.