The chip factory and slot vocabulary, the evergreen docs chips, the per-page slot
table and the signal chips each get their own module; registry.ts is now the
public face that re-exports them. A pure move.
dashboardAgentWatchChecks.ts keeps the dispatcher, the failure envelope and the
previous-facts reader, and re-exports the run, queue, error and health families
plus the shared reader contract. A pure move; every import path is unchanged.
watch-tick.ts keeps the two task definitions and the per-watch check; the wake
delivery, the condition lifecycle, the group tick and the db/session/callback
adapters each move to their own module. A pure move: watch-tick.ts re-exports
everything it exported before.
tools.ts now only assembles ready adapters, in the same frozen key order: the
HTTP/JWT client, result curation, the docs client, the source-read ledger,
evidence canonicalisation, investigation persistence, and the api/navigation/
watch/alert tool groups each own their own module. A pure move.
The prompt-prefix fingerprints are unchanged, which is what keeps the head-start
and agent prefixes byte-identical.
The docs claimed every turn is judged; the sampled rate, the redaction rule, the code-mode
skip, the opt-out and the retention period are now written down in one place.
Payloads, outputs, query rows and file contents are replaced by their shape before a turn
reaches the judge, and the row now keeps the derived verdict only.
The wake narration and watch investigation cases move verbatim to
watch-actions.test.ts; the scaffolding both test files use moves to
test-support.ts. Same 146 passed / 2 skipped.
Pure move. The wake narration and the consented investigation now live in
watch-actions.ts, reached from one call in onAction; the runtime they share with
the agent's own hooks moves to agent-runtime.ts. Every export is re-exported from
dashboard-agent.ts, so no import path changes.
The gallery pages, their fixtures and the screenshot script only exist to
look at the agent UI, so they ship separately. The fixtures two unit tests
read — the report view models and the page contexts — stay here.
Head start passed `system` as a bare string, so Anthropic neither wrote nor read
the cache: the ~18.4k-token prefix was billed in full and the agent's step 2 then
paid for a fresh write. Both breakpoints now carry a 1-hour TTL.
The two prefixes were also not identical — `get_run` sat in a different position in
the agent's tool set than in the canonical schema-only one, so they could never have
shared a cache. Reordered, and a fingerprint over system text + tool definitions is
logged per model call alongside the provider's cache-write, cache-read and uncached
input token counts, so a future drift is visible instead of silent.
A single read could put ~65k tokens in the transcript, paid for on every later
turn. The truncation notice points at the line range, and the range is now applied
before the cap so a read past the ceiling returns the lines asked for.
render_view and get_report both return the panel's view model. Their toModelOutput
now gives the model an acknowledgement and the graded summary respectively, while
the client still receives the blocks and the full report. Errors pass through
verbatim, so the fix-and-render-again loop is unchanged.
The per-turn LLM judge was billing every turn, and an unparseable
DASHBOARD_AGENT_EVAL_SAMPLE_RATE fell back to full rate too. Nothing reads
chat_turn_evals yet, so 10% loses no signal we consume.
Katia Bulatova