Once this is merged, oxlint is at a pretty sensible baseline.
**Enable `no-unused-vars`, `typescript/consistent-type-imports`, and
`import/no-duplicates` lint rules**
Turns on three previously-disabled oxlint rules across the monorepo and
fixes all violations:
- **`no-unused-vars`** – enabled as an error with standard ignore
patterns: unused function arguments are ignored by default (`args:
"none"`), variables/caught errors/destructured array elements prefixed
with `_` are allowed, and rest siblings are permitted.
- **`typescript/consistent-type-imports`** – enforced as an error; all
type-only imports now use the `import type` syntax.
- **`import/no-duplicates`** – enforced as an error; duplicate import
statements from the same module have been merged.
The remaining commits clean up the violations found across the codebase:
removing unused variables/imports/type aliases, adding `_` prefixes to
intentionally unused bindings, fixing duplicate imports, and converting
value imports to `import type` where appropriate.
Fixes#4046
## Problem
When a single run creates more than 1000 idempotency keys (e.g. a large
batch trigger where each item calls `idempotencyKeys.create()`), the
original key and scope metadata is silently dropped for all but the most
recent 1000 keys.
`idempotencyKeys.create()` returns a plain 64-char hash and stores the
`{ key, scope }` mapping in an in-process catalog keyed by that hash.
That catalog was a fixed-size **LRU capped at 1000 entries**. Once a run
creates more than 1000 keys, the earliest mappings are evicted, so when
the SDK later looks them up to attach `idempotencyKeyOptions` to the
trigger call, it finds nothing and sends `undefined`. The affected runs
then:
- report `ctx.run.idempotencyKey` as the raw hash instead of the
user-provided key
- have no `idempotencyKeyScope`
- show empty `idempotency_key` / `idempotency_key_scope` in the
dashboard and analytics
Deduplication still works (the hash is intact); only the human-readable
metadata is lost, which makes the failure silent and hard to notice.
## Fix
- Replace the LRU catalog with an unbounded in-memory catalog, so every
key created within a run keeps its metadata regardless of how many are
created.
- Clear the catalog at each run boundary via
`resetExecutionEnvironment()` (both dev and managed workers), matching
how every other per-run manager is reset. Deployed workers reuse one
process across many runs (warm starts), so this bounds memory to a
single run's keys instead of accumulating across runs — which is the
reason the size cap existed in the first place.
## Tests
- New public-API test creates 3000 keys and asserts all of them
(including the first) retain their key/scope — this fails on `main` and
passes with the fix.
- New test for the in-memory catalog covers store/retrieve/overwrite,
large-N retention (no eviction), and `clear()`.
- New test asserts the catalog is emptied after a run-boundary reset.
- Replaces the previous LRU catalog + its eviction tests.
Verified: `@trigger.dev/core` and `trigger.dev` both build; all
idempotency tests pass. Changeset added (patch).
Closes this feature request:
[https://triggerdev.featurebase.app/p/isolated-dev-sessions-for-multiple-local-trigger-dev-instances](https://triggerdev.featurebase.app/p/isolated-dev-sessions-for-multiple-local-trigger-dev-instances)
### Feature notes:
- CLI `trigger dev` works as before
- `trigger dev --branch my-branch` to create a new branch and run
against it.
- `trigger dev archive --branch my-branch` to archive (or in webapp).
- New webapp page to manage and archive dev branches, currently feature
flagged.
### Implementation details:
- No changes to data model, no backfill. `isBranchableEnvironment`
column is ignored for dev branches, we use `parentEnvironmentId IS NULL`
instead.
- `x-trigger-branch` overloaded for preview and dev branches
- New `TRIGGER_DEV_BRANCH` env var available locally.
`TRIGGER_PREVIEW_BRANCH` overloaded for child runs.
- Lots of new glue code to sanitise the branch checks.
### Rollout
- Deploy webapp/API changes (all backwards compatible)
- Manual tests on some orgs
- Deploy docs, release CLI, flip feature flag for webapp feature
### NB
- `api.v1.projects.$projectRef.environments.ts` will return
`isBranchableEnvironment: true` for all dev environments.
### Prerequisites
- [x] Typecheck will not pass until we make a new release of
`@trigger.dev/platform` and bump it here
Runners were POSTing a debug log to the supervisor for every log line -
one request per line, unbatched and unconditional. The supervisor
already has a `SEND_RUN_DEBUG_LOGS` toggle (off by default) that
discards them on receipt, but the runner fired the request regardless,
so the traffic hit the supervisor either way.
This gates the send at the source. The runner now reads
`TRIGGER_SEND_RUN_DEBUG_LOGS` (off by default, injected by the
supervisor from its existing `SEND_RUN_DEBUG_LOGS` setting) and skips
the POST entirely when disabled. Local log output is unchanged. Dev runs
use a separate path and are unaffected.
## Summary
Adds a short-lived, delegated token (`tr_uat_...`) that authenticates
against the API as a user without handing out a long-lived personal
access token. You mint one from a PAT, optionally narrow it to a set of
scopes, and give it a lifetime; the API then treats requests as that
user, subject to their role.
`trigger.dev mint-token` is the entry point (it uses your stored PAT):
```bash
UAT=$(trigger.dev mint-token --ttl 3600 --cap read:runs)
```
The token works anywhere a PAT does for user-level endpoints, and can be
exchanged for an environment JWT at `POST
/api/v1/projects/:ref/:env/jwt` to reach environment-scoped data (the
same exchange a PAT supports).
## How it works
A user-actor token is a short-lived JWT verified by a new first-class
`authenticateUserActor` method on the RBAC plugin. Self-hosters get a
built-in fallback; role-aware enforcement comes from the plugin.
Effective permissions are the intersection of the user's role and the
token's optional scope cap, so a token is only ever narrower than the
user, never broader.
Minting is restricted to personal access tokens (a token can't mint
another one, and an environment key can't mint one). Tokens default to a
1 hour lifetime (max 365 days). When exchanged for an environment JWT,
the user is stamped on it for attribution and the scope cap is carried
through.
## Summary
Three improvements to the SDK-bundled agent skills (follow-up to the
skills installer):
- **`trigger-` namespace.** The installed skills (`authoring-tasks`,
`getting-started`, …) had generic names that collide with unrelated
skills in a shared agent skills directory. They're now prefixed —
`trigger-authoring-tasks`, `trigger-getting-started`, etc. — matching
the convention the public skills repo already uses.
- **New `trigger-cost-savings` skill.** An MCP-driven cost audit:
right-sizes machines, flags missing `maxDuration`, spots sequential
triggers that could batch, and reviews schedule frequency, using
`list_runs` / `get_run_details` for live analysis.
- **Bundle the full docs.** `@trigger.dev/sdk` now bundles the entire
"Documentation" section of the docs (157 pages) instead of a curated
55-page subset, so an agent has the complete, version-pinned reference
in `node_modules`.
## How the bundling works
`scripts/bundleSdkDocs.ts` now reads `docs/docs.json`, walks the
"Documentation" dropdown, and copies every page under it into the SDK.
The set tracks the docs navigation automatically — add a page to the nav
and it ships, no skill edits needed. The API reference and Guides &
examples dropdowns are intentionally excluded. A skill's `sources:`
frontmatter is now informational only.
The dropped idea of a dedicated `trigger-config` skill is replaced by
references to the bundled build-extension docs (`config/extensions/*`)
from the `trigger-authoring-tasks` config section and the chat-agent
skills.
## Summary
Running `trigger.dev dev` before setting up a project crashed with a raw
`Cannot find matching package.json` stack trace from a transitive
dependency, instead of telling the user what to do next. It happens
whenever `dev` (or `update`) runs in a directory with no `package.json`
in it or any parent directory, for example right after creating an empty
project folder, or when `init` was exited before it scaffolded anything.
The CLI now detects the missing project and prints actionable guidance
pointing at `init`.
## Fix
`dev` runs an embedded package-version check before it loads any project
config. That check resolved `package.json` through a helper that throws
when nothing is found up the tree, and nothing caught it. It is now
wrapped, so a missing `package.json` produces a clear "run init" message
and a clean exit.
The config loader had the same latent crash on the `--skip-update-check`
path. Its resolvers for `package.json`, the lockfile, and the workspace
root all ran before the friendly "couldn't find your trigger.config.ts"
check, so any of them throwing masked it. That check now runs first and
short-circuits before the resolvers touch the filesystem.
Verified live: in an empty directory, `dev`, `dev --skip-update-check`,
and `update` all print a "run init" message and exit cleanly; in a
configured project, `dev` still resolves config and boots normally.
## What
`dev` and `deploy` now fail with a clear error when two tasks are
defined with the same id — including across task types (e.g. a scheduled
task and a regular task sharing an id).
## Why
Tasks are registered into the resource catalog keyed by id, so a second
definition with the same id silently overwrote the first. One of the
tasks would just vanish from the worker with no warning — easy to miss,
hard to debug. (Any earlier duplicate-id check ran against the
post-registration task list, which is already de-duplicated, so it never
actually fired.)
## How
- **Detect at registration** (`@trigger.dev/core`):
`StandardResourceCatalog` records a collision when a task id is
registered more than once, capturing the files involved — the only point
where both definitions are visible before the id-keyed map collapses
them. Exposed via `listTaskIdCollisions()`.
- **Fail indexing** (`trigger.dev` CLI): both index workers report
collisions via a new `TASKS_FAILED_TO_INDEX` message;
`indexWorkerManifest` rejects with a new `DuplicateTaskIdsError`. `dev`
renders a dedicated error (offending ids + files + docs link); `deploy`
fails with the same message. Runtime worker boot is unaffected — it
never reads the collisions.
- **Server-side backstop** (webapp): background-worker registration also
rejects duplicate ids with a clear `ServiceValidationError`, so
duplicates are caught even from an older CLI.
## Testing
- Unit tests for collision collection in the catalog and for the
error-message formatting (standard, same-file, and 3+-definition cases).
- Verified end to end against a local webapp: a project with a regular
task and a scheduled task sharing an id now fails `dev` with the
dedicated error; a project with distinct ids still starts normally.
## Changeset
Patch for `@trigger.dev/core` and `trigger.dev`.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
## Summary
`trigger init` now sets up your AI coding assistant as part of project
setup. Instead of the old either/or "MCP or CLI" prompt, it offers the
MCP server and agent skills together, then asks whether to scaffold with
the CLI or let your assistant do it.
A new `getting-started` agent skill backs that hand-off: it teaches the
assistant the bootstrap recipe (install the SDK, write
`trigger.config.ts`, scaffold a first task, wire tsconfig/gitignore, run
`trigger dev`) and is explicit about the two steps that genuinely need a
human (`trigger login` and copying the DEV secret key from the
dashboard). It ships in the CLI alongside the existing skills,
version-matched to your SDK.
Prompt-once gating is shared, so opting in or out during `init` means
`trigger dev` won't ask about skills again.
## Summary
`trigger skills` installs Trigger.dev agent skills into your coding
agent so it knows how to write Trigger.dev code: tasks, schedules,
realtime, and `chat.agent` AI agents. The skills are `SKILL.md` files
(the open Agent Skills format) bundled with the CLI and copied into each
tool's native skills directory (Claude Code, Cursor, GitHub Copilot, and
Codex / `AGENTS.md`), version-matched to the CLI you run. `trigger dev`
offers to install them on first run, and a one-line always-on pointer is
written into your `CLAUDE.md` / Cursor rules / etc. so the agent always
knows which skills are available and loads the right one on demand.
This replaces the old `install-rules` command, which stays as an alias.
Four skills ship to start: `authoring-tasks`, `realtime-and-frontend`,
`authoring-chat-agent`, and `chat-agent-advanced`.
## Summary
The Trigger.dev MCP server told the AI agent to wait for the run to
complete after every `trigger_task` call. The agent followed that
instruction even when the user only wanted to fire-and-forget, which
burned tokens polling runs nobody needed to block on and made responses
less clear.
Waiting is now opt-in. After triggering, the response tells the agent
the run is executing in the background and to only wait if the user
asked it to (for example "trigger and then wait for it to finish"). The
`trigger_task` tool description is updated to match. The
`wait_for_run_to_complete` tool itself is unchanged, so explicit waits
still work.
## Summary
`trigger.dev dev` was silently dropping registered `chat.agent` skills
for any project whose task files read `process.env` at module top level
— e.g. a third-party SDK client initialized at import. The agent would
boot fine, but `skill.local()` failed at runtime with `ENOENT` because
the skill folder was never copied into `.trigger/skills/`.
## Design
The CLI ran two indexer passes in dev: the worker's own indexer (with
the full env it eventually executes tasks in), and a separate
skill-discovery indexer with only the CLI process's env. Top-level reads
of vars like `TRIGGER_API_URL` imported cleanly in the worker pass and
threw in the skill pass — the latter caught the error, warned, and
skipped skill copying. Failure was silent enough that `skill.local()`
only surfaced it at task runtime.
The skill registry is already part of the worker manifest. This PR drops
the duplicate pass and copies skill folders from that manifest after the
worker initializes. One indexer instead of two; a bad `SKILL.md` now
surfaces as a startup error instead of silently disappearing skills.
Deploy is unaffected — its skill discovery uses the project's
environment variables (fetched via the API, which fills in
`TRIGGER_API_URL` etc.), so the dev failure mode doesn't reach there.
## Test plan
- [x] New `references/agent-skills` reference project with
`skills.define` + a task that calls `skill.local()` and runs a bundled
script
- [x] On `main`, adding a top-level
`process.env.TRIGGER_API_URL!.includes(...)` read in any task file
reproduces the symptom: warning at dev startup, no `.trigger/skills/`
folder, `skill.local()` fails with ENOENT
- [x] On this branch, same project boots clean and `skill.local()` works
end-to-end
- [x] Deploy still works end-to-end with the new reference project
On a warm worker process, a task whose `task()` definition is loaded via
`await import(...)` from inside another task's `run()` could end up
permanently missing from the catalog: the `task()` call fired with no
`_currentFileContext` set, `registerTaskMetadata` silently returned, and
Node's ESM module cache then blocked the worker's setContext + re-import
recovery from ever firing the call again. Subsequent runs of that task
on the same warm process failed with `COULD_NOT_FIND_EXECUTOR` until the
process hit `maxExecutionsPerProcess` and exited.
All five of these had to coincide on the same worker for the bug to
surface:
1. `processKeepAlive` enabled (so catalog state survives across runs).
2. A `run()` function (or lifecycle hook) does `await import(...)`.
3. The import's transitive static graph reaches a `task()` /
`schemaTask()` call.
4. The task containing the dynamic import is the **first** task to run
on a given warm worker process — so the dropped `task()` calls fire on
this process for the first time, are silently dropped, and Node's module
cache locks the wrong outcome in.
5. A subsequent run for one of the dropped task ids lands on the same
warm worker before it recycles.
The runtime workers now set a sentinel file context (`<no-context>`)
around the `executor.execute(...)` call, so `task()` invocations firing
during a run register normally. The catalog detects the sentinel and
emits a one-time `console.warn` per task id so the pattern stays visible
without spamming. The indexer never sets this context, so deploy-time
behavior is unchanged.
Repro is `references/hello-world/src/trigger/dynamicImportRepro*.ts`.
Verified end-to-end against a deployed image with firestarter
warm-starts on: pre-fix saw `COULD_NOT_FIND_EXECUTOR` on children that
landed on the parent-poisoned worker; post-fix all 23/23 runs succeeded
and the warning surfaces in the parent's run trace.
## Summary
Long-running chat agents were filling `session.out` forever — every
`chat.agent` turn appended to the same S2 stream with no trim, and the
Sessions dashboard re-streamed the entire history from `seq_num=0` on
every page load. After this change the agent appends an S2 `trim`
command record after each `trigger:turn-complete`, pointing back at the
previous turn-complete's seq_num. `session.out` stays roughly one turn
long at steady state, regardless of session age.
`trigger:turn-complete` and `trigger:upgrade-required` move from
`chunk.type`-shaped data records into header-form control records under
a uniform `trigger-control` namespace. Built-in transports
(`TriggerChatTransport`, `AgentChat`, the dashboard's `AgentView`)
handle the new shape transparently. Custom transports need a one-line
filter on the `trigger-control` header — see the rewritten "Records on
session.out" section in the client-protocol docs.
The Sessions detail page in the dashboard fetches the agent's per-turn
S3 snapshot via a presigned URL and seeds the transcript view, then
SSE-tails from the snapshot's `lastOutEventId`. Bandwidth and
time-to-first-render scale with unread turns instead of session
lifetime.
Resume contract is now explicit: single-turn-boundary resume always
works (the prior turn-complete is still on the stream), the S2 trim is
eventually consistent over 10-60s, and multi-turn-away resume falls back
to a snapshot reload.
## Summary
Local self-hosted deploys (`trigger deploy --local-build --push
--builder orbstack` or any other buildx setup using the **docker**
driver) fail at the push step with:
```
ERROR: failed to build: failed to solve:
exporter option "rewrite-timestamp" conflicts with "unpack"
```
The docker driver auto-enables `unpack=true` when pushing, and that's
incompatible with `rewrite-timestamp` (which the CLI sets for
reproducible-build hashing).
Adds a simple env-var opt-out so contributors can keep using their
default builder. The flag is only read by the local-build code path;
remote/cloud builds are unaffected.
```bash
TRIGGER_BUILD_SKIP_REWRITE_TIMESTAMP=1 \
pnpm exec trigger deploy --profile default --local-build --push --builder orbstack
```
The trade-off: skipping `rewrite-timestamp` means layer timestamps
reflect actual build time, so two identical builds produce different
layer hashes. Fine for a local-dev registry; the only real consumer of
timestamp-stability is registry-layer cache hit rates.
## Test plan
- [x] Manual: ran `trigger deploy --profile default --local-build --push
--builder orbstack` against the localhost webapp + a local Docker
registry on port 5001 — first failed with the rewrite-timestamp/unpack
error, then succeeded after setting
`TRIGGER_BUILD_SKIP_REWRITE_TIMESTAMP=1`.
- [x] Full chat.agent smoke sweep (15 tests, including suspend/resume,
deepResearch subtask, AgentChat orchestrator) against the deployed image
— all pass.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
## Summary
Enables shipping `X.Y.Z-rc.N` prereleases of `@trigger.dev/*` via
changesets pre mode. RCs publish under the `rc` npm dist-tag, never
claim `latest`, and don't trigger marketing-site changelog PRs. The
plumbing is hyphen-in-version detection in `release.yml` — no separate
workflow, no opt-in flag at publish time.
Validated end-to-end against a sandbox repo (real npm publishes, Docker
builds, Helm chart pushes, GitHub releases) before porting back. Full RC
lifecycle tested: pre enter → rc.0 → iterate to rc.1 → pre exit →
stable. Plus interaction with the existing release-branch hotfix flow.
## What changes
### `release.yml`
- New `is_prerelease` output (hyphen-in-version)
- GitHub release adds `--prerelease` flag for RC publishes (Pre-release
badge, not Latest)
- `dispatch-changelog` job gated on `is_prerelease != 'true'` — no
marketing-site PR per RC
### Docker workflows
- Removes the `:v4-beta` floating tag entirely from `publish-webapp.yml`
and `publish-worker-v4.yml`. v4 is GA; the tag is a misnomer and is
already inconsistent with the npm side (npm `v4-beta` dist-tag was
frozen at 4.0.4 months ago while Docker `:v4-beta` kept bumping).
Self-hosters should pin to a versioned tag going forward — the last
value of `:v4-beta` stays frozen wherever it currently points.
### CLI version-check fix
(`packages/cli-v3/src/utilities/initialBanner.ts`)
Switches the "new version available" comparison from JavaScript
`localeCompare` to `semver.lt`. The old comparison handled `X.Y.Z-rc.N`
vs `X.Y.Z` incorrectly — a user on `4.5.0-rc.0` would never be prompted
to upgrade once `4.5.0` stable shipped (lex order put the prerelease
ahead of the bare version). Real semver gets this right.
Stable users were never affected: the check queries the `@latest`
dist-tag, which by convention never points at a prerelease.
## How an RC actually publishes after this
1. `pnpm exec changeset pre enter rc` on main, push the `pre.json`
2. Bot regenerates the release PR as `chore: release v<X.Y.Z>-rc.0`
3. Merge → `release.yml` runs `changeset publish` which reads
`pre.json.tag` and publishes under `--tag rc`. GitHub release marked
Pre-release. No marketing-site dispatch.
4. Iterate by adding changesets normally; bot bumps to `rc.1`, `rc.2`, …
5. When ready: `pnpm exec changeset pre exit`, push, merge regenerated
PR → stable ships under `latest` and the marketing-site dispatch fires.
## Summary
Adds a Region column and Region filter (under More filters) to the runs
list dashboard, the same filter on the public runs list API
(`filter[region]`), and a matching `region` input on the MCP `list_runs`
tool. Each run's executing region is also surfaced as a new optional
`region` field on the runs list and run retrieve responses, populated
from the worker instance group's `masterQueue` identifier.
Useful when you run tasks across multiple regions and want to slice the
runs list — or your existing run-querying scripts — by where the run
actually executed.
## Design
The filter value in the URL / API is the `masterQueue` identifier (the
same string already persisted on `TaskRun` and replicated to ClickHouse
as `worker_queue`), so the query just becomes `worker_queue IN (...)`
with no server-side translation. The Region dropdown options come from a
new resource loader backed by `RegionsPresenter`, which now also exposes
`masterQueue` alongside the existing region metadata.
```ts
// public API
const runs = await runs.list({ region: ["us-east-1", "eu-west-1"] });
// each item: { id, status, ..., region?: "us-east-1" }
```
```ts
// MCP
list_runs({ environment: "prod", region: "us-east-1" })
```
Top of the chat.agent stack: a full Next.js reference project that
exercises chat.agent end-to-end, plus the CLI MCP tools that drive
agent runs from Claude Code / Cursor / etc.
references/ai-chat:
- Full Next.js app with prisma persistence, multi-chat sidebar,
per-chat model picker, debug panel, tool examples, smoke tests
- Reference tools: getCurrentTime, searchHackerNews, createGithubIssue,
PR review helpers, code sandbox
- chat-client-test orchestrator for concurrent-send stress
- references/hello-world chatAgent + triggerAndSubscribe examples
CLI MCP tooling for chat.agent:
- mcp/tools/agentChat.ts (start_agent_chat, send_agent_message,
close_agent_chat)
- mcp/tools/agents.ts + tasks.ts (list agents, agent run details)
- dev-run-controller OOM kill + taskRunProcessPool tweaks
- dev/managed entry-point hooks for skills bundling
- buildWorker + bundleSkills (agent skills support)
Includes ai-tool-helpers + mcp-agent-chat-sessions changesets, plus
the streamdown@2 patch and pnpm-lock reconciliation.
(Will be renamed to feature/ai-chat-reference-and-cli before push.)
fix(cli): preserve lastEventId after sendMessage fallback to avoid stale turn-complete replay
When a Node EventEmitter (e.g. node-redis) emits an "error" event with
no
listener attached, Node escalates it to process.on("uncaughtException")
in
the task worker. The worker reported the error via the
UNCAUGHT_EXCEPTION
IPC event but did not exit, and the supervisor-side handler in
taskRunProcess only logged the message at debug level — leaving the
run()
promise orphaned until maxDuration fired and producing empty attempts
(durationMs=0, costInCents=0).
The supervisor now rejects the in-flight attempt with an
UncaughtExceptionError and gracefully terminates the worker (preserving
the OTEL flush window) on UNCAUGHT_EXCEPTION. The attempt fails fast
with
TASK_EXECUTION_FAILED, surfacing the original error name, message, and
stack trace, and falls under the normal retry policy. This mirrors the
existing indexing-side behavior in indexWorkerManifest. Apply the same
handling to unhandled promise rejections, which Node already routes
through uncaughtException by default.
Orphaned `trigger-dev-run-worker` processes were pinning CPU at 100%
after the dev CLI exited — stuck in an uncaughtException feedback loop
where a closed IPC channel kept throwing `ERR_IPC_CHANNEL_CLOSED` back
into a handler that itself called `process.send`.
Fix:
- `ZodIpcConnection` no-ops sends when the channel is disconnected.
- Dev workers exit on `process.disconnect` instead of being re-parented
to init.
- All worker `uncaughtException` handlers route through a `safeSend`
guard so the handler can never re-enter itself.
Verified end-to-end: `kill -9` of the dev CLI now cleans up all child
workers within ~2s.
Closes the most common friction point hit while setting up a fresh
project from an agent harness: the CLI auto-opens the user's default
browser during auth and there is no supported way to skip it (the
existing `isLinuxServer()` path only triggers when `xdg-open` is missing
entirely).
`--no-browser` on `login` and `init` prints the URL and waits to be
visited from any browser. The flag threads through the embedded
`login()` call inside `init`.
While here:
- `init` now errors loudly when stdin is non-TTY without `--yes` instead
of default-and-exiting silently at the first prompt (which left the
project half-initialized: deps installed, no config or example file).
- Both commands gain an `Examples` block in `--help` rendered between
the description and the arguments/options list, so `--help | head`
surfaces the common invocations.
Other commands also call `login()` embedded and would benefit from
`--no-browser` too, but kept this PR scoped to the cases the friction
log called out.
Add TTL (time-to-live) defaults at task-level and config-level, with
precedence: per-trigger > task > config > dev default (10m).
Docs PR: #3200 (merge after packages are released)
For human reviewer:
- Check if Redis connection + code makes sense
- Check CLI methods (it's on a hotpath)
- Check DB Migrations and new tables
## ✅ Checklist
- [x] I have followed every step in the [contributing
guide](https://github.com/triggerdotdev/trigger.dev/blob/main/CONTRIBUTING.md)
- [x] The PR title follows the convention.
- [x] I ran and tested the code works
---
## Testing
Spawning new CLI / Dashboard notifications, check MVP, check if failures
not produce any problems with CLI/Dashboard
---
## Changelog
Added notifications mechanism for Dashboard and CLI
---
## Screenshots
💯
Adds an `annotations` JSONB column to task runs that captures where and
how each run was triggered.
This enables filtering and analyzing trigger origins without querying up
the run tree. Also enables making scheduling decisions based on the
trigger source, e.g., use separate affinities for scheduled runs.
Each run records:
- **triggerSource**: who initiated it (sdk, api, dashboard, cli, mcp,
schedule)
- **triggerAction**: what kind of action (trigger, replay, test)
- **rootTriggerSource**: the trigger source of the root ancestor,
propagated through the entire run
tree
- **rootScheduleId**: schedule id, in case the run tree was triggered
from a schedule
Currently the main motivation for annotations it to determine whether a
run is part of a schedule-originated tree without traversing ancestors.
### A couple of design considerations
- **Decoupled source from method**: triggerSource and triggerAction are
separate fields to avoid
combinatorial explosion (every new source × every new action)
- **Server-side first**: all annotation values are primarily determined
on the server, only a minor SDK change needed
- **Forward-compatible**: annotation fields use
`z.enum([...]).or(anyString)` so new values can be
added without breaking validation; we currently don't need an explicit
version field for annotations.
Note: `metadata` would have been a more fitting name for the db column,
as it is consistent with other tables where we store this type of
information. It is already in use to store user metadata though, so we
go with `annotations` instead.
- Full prompt management UI: list, detail, override, and version
management for AI prompts defined with `prompts.define()`
- Rich AI span inspectors for all AI SDK operations with token usage,
messages, and prompt context
- Real-time generation tracking with live polling and filtering
## Prompt management
Define prompts in your code with `prompts.define()`, then manage
versions and overrides from the dashboard without redeploying:
```typescript
import { task, prompts } from "@trigger.dev/sdk";
import { generateText } from "ai";
import { openai } from "@ai-sdk/openai";
import { z } from "zod";
const supportPrompt = prompts.define({
id: "customer-support",
model: "gpt-4o",
variables: z.object({
customerName: z.string(),
plan: z.string(),
issue: z.string(),
}),
content: `You are a support agent for Acme SaaS.
Customer: {{customerName}} ({{plan}} plan)
Issue: {{issue}}
Respond with empathy and precision.`,
});
export const supportTask = task({
id: "handle-support",
run: async (payload) => {
const resolved = await supportPrompt.resolve({
customerName: payload.name,
plan: payload.plan,
issue: payload.issue,
});
const result = await generateText({
model: openai(resolved.model ?? "gpt-4o"),
system: resolved.text,
prompt: payload.issue,
...resolved.toAISDKTelemetry(),
});
return { response: result.text };
},
});
```
The prompts list page shows each prompt with its current version, model,
override status, and a usage sparkline over the last 24 hours.
From the prompt detail page you can:
- **Create overrides** to change the prompt template or model without
redeploying. Overrides take priority over the deployed version when
`prompt.resolve()` is called.
- **Promote** any code-deployed version to be the current version
- **Browse generations** across all versions with infinite scroll and
live polling for new results
- **Filter** by version, model, operation type, and provider
- **View metrics** (total generations, avg tokens, avg cost, latency)
broken down by version
## AI span inspectors
Every AI SDK operation now gets a custom inspector in the run trace
view:
- **`ai.generateText` / `ai.streamText`** — Shows model, token usage,
cost, the full message thread (system prompt, user message, assistant
response), and linked prompt details
- **`ai.generateObject` / `ai.streamObject`** — Same as above plus the
JSON schema and structured output
- **`ai.toolCall`** — Shows tool name, call ID, and input arguments
- **`ai.embed`** — Shows model and the text being embedded
For generation spans linked to a prompt, a "Prompt" tab shows the prompt
metadata, the input variables passed to `resolve()`, and the template
content from the prompt version.
All AI span inspectors include a compact timestamp and duration header.
## Other improvements
- Resizable panel sizes now persist across page refreshes (patched
`@window-splitter/state` to fix snapshot restoration)
- Run page panels also persist their sizes
- Fixed `<div>` inside `<p>` DOM nesting warnings in span titles and
chat messages
- Added Operations and Providers filters to the AI metrics dashboard
## Screenshots
<img width="3680" height="2392" alt="CleanShot 2026-03-21 at 10 14
17@2x"
src="https://github.com/user-attachments/assets/f3e59989-a2fa-4990-a9d0-3cacda431868"
/>
<img width="3680" height="2392" alt="CleanShot 2026-03-21 at 10 15
37@2x"
src="https://github.com/user-attachments/assets/2f2d02df-2d2b-44fb-ac6f-9153f6a6c387"
/>
<img width="3680" height="2392" alt="CleanShot 2026-03-21 at 10 15
54@2x"
src="https://github.com/user-attachments/assets/baa161e0-ef91-4fa4-a55f-986b71cccdf0"
/>
- Automatic LLM cost enrichment for AI SDK spans (streamText,
generateText, generateObject) or any other spans that use semantic
gen_ai attributes with support for 145+ models
- New AI span inspector sidebar showing model, tokens, cost, messages,
tool calls, and response text
- LLM metrics dual-write to ClickHouse `llm_metrics_v1` table for
analytics
- LLM metrics built-in dashboard (unlinked at the moment)
- Provider cost fallback — uses gateway/OpenRouter reported costs from
`providerMetadata` when registry pricing is unavailable
- Prefix-stripping for gateway/OpenRouter model names (e.g.
`mistral/mistral-large-3` matches `mistral-large-3` pricing)
- Admin dashboard for managing LLM model pricing (list, create, edit,
delete, search, test pattern matching)
- Missing models detection page — queries ClickHouse for unpriced models
with sample spans and Claude Code-ready prompts for adding pricing
- AI span seed script (`pnpm run db:seed:ai-spans`) with 51 spans across
12 provider systems for local dev testing
- UI fixes: `completionTokens`/`promptTokens` aliases,
`ai.response.object` display for generateObject, cache read/write token
breakdown
## Screenshots:
<img width="1030" height="104" alt="CleanShot 2026-03-17 at 16 48 54@2x"
src="https://github.com/user-attachments/assets/bc8fccda-e48b-4d0c-bfb1-e620064e5979"
/>
<img width="1094" height="1512" alt="CleanShot 2026-03-17 at 16 49
23@2x"
src="https://github.com/user-attachments/assets/c2424569-d07e-4d67-a436-e8250043a1ee"
/>
<img width="1074" height="1412" alt="CleanShot 2026-03-17 at 16 49
18@2x"
src="https://github.com/user-attachments/assets/22342ac4-4769-45d1-a328-a24fb9a82a50"
/>
<img width="1012" height="2292" alt="CleanShot 2026-03-17 at 16 39
01@2x"
src="https://github.com/user-attachments/assets/59e327d1-6652-4293-8be0-bb8326e5fbc5"
/>
<img width="3680" height="2392" alt="CleanShot 2026-03-15 at 08 29
38@2x"
src="https://github.com/user-attachments/assets/1f77beb8-de67-495b-b890-bcdb8d7f1fe8"
/>
---------
Co-authored-by: James Ritchie <james@trigger.dev>
## Summary
Major expansion of the MCP server (14 → 25 tools), context efficiency
optimizations, new API endpoints, and a fix for the dev CLI leaking
build directories on disk.
### New MCP tools
- **Query & analytics**: `get_query_schema`, `query`, `list_dashboards`,
`run_dashboard_query` — query your data using TRQL directly from AI
assistants
- **Profile management**: `whoami`, `list_profiles`, `switch_profile` —
see and switch CLI profiles per-project (persisted to
`.trigger/mcp.json`)
- **Dev server control**: `start_dev_server`, `stop_dev_server`,
`dev_server_status` — start/stop `trigger dev` and stream build output
- **Task introspection**: `get_task_schema` — get payload schema for a
specific task (split out from `get_current_worker` to reduce context)
### New API endpoints
- `GET /api/v1/query/schema` — discover TRQL tables and columns
(server-driven, multi-table)
- `GET /api/v1/query/dashboards` — list built-in dashboard widgets and
their queries
### New features
- **`--readonly` flag** — hides write tools (`deploy`, `trigger_task`,
`cancel_run`) so agents can't make changes
- **`read:query` JWT scope** — new authorization scope for query
endpoints, with per-table granularity (`read:query:runs`,
`read:query:llm_metrics`, etc.)
- **Paginated trace output** — `get_run_details` now paginates trace
events via cursor, caching the full trace in a temp file so subsequent
pages don't re-fetch
- **MCP tool annotations** — all tools now have
`readOnlyHint`/`destructiveHint` annotations for clients that support
them
- **Project-scoped profile persistence** — `switch_profile` saves to
`.trigger/mcp.json` (gitignored), automatically loaded on next MCP
server start
### Context optimizations
- `get_query_schema` requires a table name — returns one table's schema
instead of all tables (60-80% fewer tokens)
- `get_current_worker` no longer inlines payload schemas — use
`get_task_schema` for specific tasks
- Query results formatted as text tables instead of JSON (~50% fewer
tokens for flat data)
- `cancel_run`, `list_deploys`, `list_preview_branches` formatted as
text instead of raw `JSON.stringify()`
- Schema and dashboard API responses cached (1hr and 5min respectively)
### Bug fixes
- Fixed `search_docs` failing due to renamed upstream Mintlify tool
(`SearchTriggerDev` → `search_trigger_dev`)
- Fixed `list_deploys` failing when deployments have null
`runtime`/`runtimeVersion` fields (fixes#3139)
- Fixed `list_preview_branches` crashing due to incorrect response shape
access
- Fixed `metrics` table column documented as `value` instead of
`metric_value` in query docs
- Fixed `/api/v1/query` not accepting JWT auth (added `allowJWT: true`)
### Dev CLI build directory fix
The dev CLI was leaking `build-*` directories in `.trigger/tmp/` on
every rebuild, accumulating hundreds of MB over time (842MB observed).
Three layers of protection added:
1. **During session**: deprecated workers are pruned (capped at 2
retained) when no active runs reference them, preventing unbounded
accumulation
2. **On SIGKILL/crash**: the watchdog process now cleans up
`.trigger/tmp/` when it detects the parent CLI was killed
3. **On next startup**: existing `clearTmpDirs()` wipes any remaining
orphans
## Test plan
- [ ] `pnpm run mcp:smoke` — 17 automated smoke tests for all read-only
MCP tools
- [ ] `pnpm run mcp:test list` — verify 25 tools registered (21 in
`--readonly` mode)
- [ ] `pnpm run mcp:test --readonly list` — verify write tools hidden
- [ ] Manual: start dev server, trigger task, rebuild multiple times,
verify build dirs stay capped at 4
- [ ] Manual: SIGKILL the dev CLI, verify watchdog cleans up
`.trigger/tmp/`
- [ ] Verify new API endpoints return correct data: `GET
/api/v1/query/schema`, `GET /api/v1/query/dashboards`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
When the dev CLI exits (e.g. ctrl+c via pnpm), runs that were
mid-execution
previously stayed stuck in EXECUTING status for up to 5 minutes until
the
heartbeat timeout fired. Now they are cancelled within seconds.
The dev CLI spawns a lightweight detached watchdog process at startup.
The
watchdog monitors the CLI process ID and, when it detects the CLI has
exited,
calls a new POST /engine/v1/dev/disconnect endpoint to cancel all
in-flight
runs immediately (skipping PENDING_CANCEL since the worker is known to
be dead).
Watchdog design:
- Fully detached (detached: true, stdio: ignore, unref()) so it survives
even when pnpm sends SIGKILL to the process tree
- Active run IDs maintained via atomic file write
(.trigger/active-runs.json)
- Single-instance guarantee via PID file (.trigger/watchdog.pid)
- Safety timeout: exits after 24 hours to prevent zombie processes
- On clean shutdown, the watchdog is killed (no disconnect needed)
Disconnect endpoint:
- Rate-limited: 5 calls/min per environment
- Capped at 500 runs per call
- Small counts (<= 25): cancelled inline with pMap concurrency 10
- Large counts: delegated to the bulk action system
- Uses finalizeRun: true to skip PENDING_CANCEL and go straight to
FINISHED
Run engine change:
- cancelRun() now respects finalizeRun when the run is in EXECUTING
status,
skipping the PENDING_CANCEL waiting state and going directly to FINISHED
Input streams enable sending typed data to executing tasks from external
callers — backends, frontends, or other tasks. This unlocks interactive
use cases like approval UIs, cancel buttons, chat interfaces, and
human-in-the-loop AI workflows where the task needs to receive data
while running.
Three consumption patterns inside a task:
* `.wait()` — Suspend the task until data arrives (process freed, most
efficient)
* `.once()` — Wait for the next message (process stays alive)
* `.on()` — Subscribe to a continuous stream of messages
One send pattern from outside:
* `.send(runId, data)` — Send typed data to a specific run's input
stream
## User-facing API
### Define a typed input stream
```ts
import { streams, task } from "@trigger.dev/sdk";
const approval = streams.input<{ approved: boolean; reviewer: string }>({ id: "approval" });
```
### Consume inside a task
```ts
export const myTask = task({
id: "my-task",
run: async () => {
// Pattern 1: Suspend until data arrives (most efficient — frees the process)
const result = await approval.wait({ timeout: "5m" });
// Pattern 2: Wait for next message (process stays alive)
const data = await approval.once().unwrap();
// Pattern 3: Subscribe to multiple messages
approval.on((data) => { /* handle each message */ });
},
});
```
### Send from outside
```ts
// From a backend (using secret API key)
await approval.send(runId, { approved: true, reviewer: "alice" });
// From a frontend (using public JWT token from trigger response)
const { send } = useInputStreamSend("approval", runId, { accessToken });
send({ approved: true, reviewer: "alice" });
```
---------
Co-authored-by: Claude <noreply@anthropic.com>
- Adds an end-to-end OTEL metrics pipeline: task workers collect and
export metrics via OpenTelemetry, the webapp ingests them into
ClickHouse, and they're queryable through the existing dashboard query
engine
- Workers emit process CPU/memory metrics (via
`@opentelemetry/host-metrics`) and Node.js runtime metrics (event loop
utilization, event loop delay, heap usage)
- Users can create custom metrics in their tasks via
`otel.metrics.getMeter()` from `@trigger.dev/sdk`
- Metrics are automatically tagged with run context (run ID, task slug,
machine, worker version) so they can be sliced per-run, per-task, or
per-machine
- The TSQL query engine gains metrics table support with typed attribute
columns, `prettyFormat()` for human-readable values, and per-schema time
bucket thresholds
- Includes reference tasks
(`references/hello-world/src/trigger/metrics.ts`) demonstrating
CPU-intensive, memory-ramp, bursty workload, and custom metrics patterns
## What changed
### Metrics collection (packages/core, packages/cli-v3)
- **Metrics export pipeline** — `TracingSDK` now sets up a
`MeterProvider` with a `PeriodicExportingMetricReader` that chains
through `TaskContextMetricExporter` (adds run context attributes) and
`BufferingMetricExporter` (batches exports to reduce overhead)
- **Host metrics** — Enabled `@opentelemetry/host-metrics` for process
CPU, memory, and system-level metrics
- **Node.js runtime metrics** — New `nodejsRuntimeMetrics.ts` module
using `performance.eventLoopUtilization()`, `monitorEventLoopDelay()`,
and `process.memoryUsage()` to emit 6 observable gauges
- File system and diskio metrics
- **Custom metrics** — Exposed `otel.metrics` from `@trigger.dev/sdk` so
users can create counters, histograms, and gauges in their tasks
- **Machine ID** — Stable per-worker machine identifier for grouping
metrics
- **Dev worker** — Drops `system.*` metrics to reduce noise, keeps
sending metrics between runs in warm workers
### Metrics ingestion (apps/webapp)
- **OTEL endpoint** — `otel.v1.metrics.ts` accepts OTEL metric export
requests (JSON and protobuf), converts to ClickHouse rows
- **ClickHouse schema** — `017_create_metrics_v1.sql` with 10-second
aggregation buckets, JSON attributes column, 60-day TTLs
### Query engine (internal-packages/tsql, apps/webapp)
- **Metrics query schema** — Typed columns for metric attributes
(`task_identifier`, `run_id`, `machine_name`, `worker_version`, etc.)
extracted from the JSON attributes column
- **`prettyFormat()`** — TSQL function that annotates columns with
format hints (`bytes`, `percent`, `durationSeconds`) for frontend
rendering without changing the underlying data
- **Per-schema time buckets** — Different tables can define their own
time bucket thresholds (metrics uses tighter intervals than runs)
- **AI query integration** — The AI query service knows about the
metrics table and can generate metric queries
- **Chart improvements** — Better formatting for byte values,
percentages, and durations in charts and tables
### Reference project
- **`references/hello-world/src/trigger/metrics.ts`** — 6 example tasks:
`cpu-intensive`, `memory-ramp`, `bursty-workload`, `sustained-workload`,
`concurrent-load`, `custom-metrics`
## Test plan
- [ ] Build all packages and webapp
- [ ] Start dev worker with hello-world reference project
- [ ] Run `cpu-intensive`, `memory-ramp`, and `custom-metrics` tasks
- [ ] Verify metrics in ClickHouse: `SELECT DISTINCT metric_name FROM
metrics_v1`
- [ ] Query via dashboard AI: "show me CPU utilization over time"
- [ ] Verify `prettyFormat` renders correctly in chart tooltips and
table cells
- [ ] Confirm dev worker drops `system.*` metrics but keeps `process.*`
and `nodejs.*`
Fixes an issue introduced in #3024.
The behavior for local builds in older CLI versions relies on
`externalBuildData` to be defined to distinguish from the self-hosting
local build path, even though it doesn't actually use the token.
## Summary
- Adds an optional `timeoutInSeconds` parameter (default 60s) to the
`wait_for_run_to_complete` MCP tool
- If the run doesn't complete within the timeout, returns the current
run state instead of blocking indefinitely
- Uses `AbortSignal.timeout()` combined with the existing MCP signal
Fixes#3032
## Summary
- When a child process crashes and a retry (`RETRY_IMMEDIATELY`) is
attempted on the same `TaskRunProcess`, `execute()` hangs forever
because the IPC send is silently skipped and the attempt promise can
never resolve
- This caused runner pods to stay up indefinitely with no heartbeats or
polls
- Fix: reject the attempt promise immediately when the child is not
connected, so the controller can proceed to warm start or exit
## Test plan
- [x] Added `taskRunProcess.test.ts` — verifies `execute()` rejects
promptly instead of hanging when the child process is dead
- [x] Deploy and verify no more stuck runner pods accumulate over time
This fixes a regression introduced in #2778 - stable sort is required
for deterministic builds, but we can safely preserve order for the user
package.json during package updates
Adds support for **debounced task runs** - when triggering a task with a
debounce key, subsequent triggers with the same key will reschedule the
existing delayed run instead of creating new runs. This continues until
no new triggers occur within the delay window.
## Usage
```typescript
await myTask.trigger({ userId: "123" }, {
debounce: {
key: "user-123-update",
delay: "5s",
mode: "leading", // default
}
});
```
- **key**: Scoped to the task identifier
- **delay**: How long to wait before executing (supports duration
strings like `"5s"`, `"1m"`)
- **mode**: Either `"leading"` or `"trailing"`. Leading debounce will
use the payload and options from the first run created with the debounce
key. Trailing will use payload and options from the last run.
### "trailing" mode overrides
When using `mode: "trailing"` with debounce, the following options are
updated from the **last** trigger:
- **`payload`** - The task input data
- **`metadata`** - Run metadata
- **`tags`** - Run tags (replaces existing tags)
- **`maxAttempts`** - Maximum retry attempts
- **`maxDuration`** - Maximum compute time
- **`machine`** - Machine preset (cpu/memory)
## Behavior
- **First run wins**: The first trigger creates the run, subsequent
triggers push its execution time later
- **Idempotency keys take precedence**: If both are specified,
idempotency is checked first
- **Max duration**: Configurable via `DEBOUNCE_MAX_DURATION_MS` env var
(default: 10 minutes)
Works with `triggerAndWait` - parent runs correctly block on the
debounced run.
This PR makes our image builds deterministic and reproducible by
ensuring that identical source code always produces the same image
layers and image digest. This means that deployments where nothing has
changed will no longer invalidate the image cache in our worker cluster
nodes, thus avoid making the cold starts for runs worse.
**Context**
New deployments currently increase the cold start times for runs, as
they generate a new image which needs to be pulled in the worker cluster
where runs are executed. It happens also when the source code for the
deployment has not changed due to non-deterministic steps in our build
system. This addresses the latter issue by making builds reproducible.
**Main changes**
- Avoided baking `TRIGGER_DEPLOYMENT_ID` and
`TRIGGER_DEPLOYMENT_VERSION` in the image, we now pass these via the
supervisor instead.
- Used `json-stable-stringify` for consistent key ordering in the files
we generate for the build, e.g., `package.json`, `build.json`,
`index.json`.
- Removed `metafile.json` from the image contents as it is not actually
used in the container. This is only relevant for the `analyze` command.
- Added `SOURCE_DATE_EPOCH=0` and `rewrite-timestamp=true` to Docker
builds to normalize file timestamps.
- Removed some `timings` and `outputHashes` from build outputs and
manifests.
The builds are now reproducible for both native build server and Depot
paths. This should also lead to better image layer cache reuse in
general.
This will speed up ice cold starts (*) for two reasons:
- better compression ratio
- faster decompression
This is a minor release because zstd compression will now be enabled by
default for all deployments.
(*) ice cold starts happen when deploy images are not cached on the
worker node yet. These cold start durations are highly dependent on
image size and as it turns out, also the type of compression used.
<!-- CURSOR_SUMMARY -->
> [!NOTE]
> Centralizes SIGTERM handling in `DevSupervisor` and removes per-run
SIGTERM listeners in `DevRunController` to avoid
MaxListenersExceededWarning under high concurrency.
>
> - **Dev runtime**:
> - **SIGTERM handling**: Add centralized handler in
`packages/cli-v3/src/dev/devSupervisor.ts` to gracefully stop all run
controllers; unregisters on `shutdown()`.
> - **Cleanup**: Remove per-controller `SIGTERM` listener and handler
from `packages/cli-v3/src/entryPoints/dev-run-controller.ts` to reduce
event listeners and warnings.
> - **Changeset**: Add patch note in
`.changeset/fuzzy-ghosts-admire.md`.
>
> <sup>Written by [Cursor
Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit
5ad2f5341829cebf6fd37a3c616a2db5e4ad936a. This will update automatically
on new commits. Configure
[here](https://cursor.com/dashboard?tab=bugbot).</sup>
<!-- /CURSOR_SUMMARY -->
This PR applies a small change to the deployments table to keep track
of:
- where the deployment was triggered from
- build server metadata, if the build server was involved
Chris Arderne