[objectstack-fleet]

Requested by Elon · project thread

Fixes #25988

Clause-②: no
Metadata-protocol: none

The claim is 6411275777 (seat domain:devx#1, card thread session_01H98bQgxPLVbpyR6Zsu6SUW, branch claude/project-thread-vi44lw, platform-assigned).

What changes

main's Dogfood Regression Gate (3/3) has been red on every run that runs it since 69fa23d6 (#25956), on one assertion of the authz blind-spot census:

control "handleI18nRequest" moved in packages/runtime/src/domains/i18n.ts: expected 4 to be 3

The census is a measurement pin, so the fourth occurrence was read before it was recorded. One file changes: packages/qa/dogfood/test/authz-probe-blind-spot.census.ts, the packages/runtime/src/domains/i18n.ts row. The handleI18nRequest control goes from 3 to 4, and the row's comment names the fourth occurrence and why reach is unchanged. No other row, no total and no derivation rule moves.

The fourth occurrence, read

packages/runtime/src/domains/i18n.ts at 05f7758f holds handleI18nRequest four times:

line	what it is	since
20	file docblock, a {@link handleI18nRequest}	before
73	TRANSLATIONS_CACHE_CONTROL's docblock, (see the floor in {@link handleI18nRequest})	69fa23d6
150	the DomainRoute call site in createI18nDomain	before
155	export async function handleI18nRequest(	before

69fa23d6 (the translations validator, Part of #25785) added three module-private helpers to that file (requestHeader, ifNoneMatchHits, translationsAnswer) and the one constant. None of them is an export async function handle*Request entry point, it added no DomainRoute prefix, and it added no shouldDenyAnonymous( call. The anonymous floor is still the handler's first statement. So the row's population, reach and gate control are unchanged at 1 / 1 / 1, and the row's own population and reach test stayed green on main the whole time. The only reading that moved is the handler-name control, by a docblock link. This is a count change, not a census row change, and nothing here needs a new probe.

Sweep of the rest of the census

• At this branch's base 05f7758f the census file reads 1 failed | 46 passed (the same i18n assertion, nothing else), and with this change 47 passed. So no other control, row, total or probe-table reading has moved since 69fa23d6 or with any later merge up to 05f7758f.
• origin/main has since moved to a4eae064 (4 commits). In the census-read source directories they touch only two test files (packages/runtime/src/domains/mcp-package-object-discovery.test.ts, packages/runtime/src/stored-metadata-reader-contexts.pin.test.ts), and the census reads neither, so the reading holds there too.

Verification

All at head ed1c5b39, under scripts/pm/os-verify-lock.sh.

• Census before the change (05f7758f, pnpm exec vitest run test/authz-probe-blind-spot.test.ts in packages/qa/dogfood): Tests 1 failed | 46 passed (47), AssertionError: control "handleI18nRequest" moved in packages/runtime/src/domains/i18n.ts: expected 4 to be 3.
• Closure build, the CI step's command at --concurrency=2: pnpm turbo run build --filter=@objectstack/dogfood: Tasks: 67 successful, 67 total, VERDICT command-exit 0.
• Dogfood shard 3/3, the CI leg's carrier and command: OS_TEST_SHARD=3/3 node scripts/run-with-stall-guard.mjs --log dogfood.log --stall-minutes 10 -- pnpm turbo run test --filter=@objectstack/dogfood:
  • Test Files 97 passed | 2 skipped (99)
  • Tests 990 passed | 14 skipped (1004)
  • authz-probe-blind-spot.test.ts (47 tests) passed inside the shard
  • VERDICT command-exit 0
  • node scripts/check-test-completeness.mjs: OK (1 package(s), 1004 test(s) declared and all 1004 accounted for)
• authz-conformance.test.ts + the census (conformance sits in another shard, so it was run on its own): Test Files 2 passed (2), Tests 108 passed (108), VERDICT command-exit 0.
• pnpm --filter @objectstack/dogfood typecheck: VERDICT command-exit 0.
• pnpm exec eslint --no-inline-config on the changed file: exit 0.
• Derived gates. node scripts/pm/dispatch-gates.mjs --commands derives 52 families on the 1 changed path. 51 exit 0. pnpm check:dual-build-cjs-loads is NOT MEASURED: it exited 3 (PREREQUISITE NOT MET) because it needs every package's dist/ and only the dogfood closure was built. CI's Lint lane owns it. dispatch-gates.mjs --ran: 52 derived, 51 run, 1 NOT-MEASURED, 0 UNRUN.
• node scripts/pm/check-governed-merges.mjs --branch HEAD: NOT governed, 16 changed lines.
• No changeset: @objectstack/dogfood is private, and the diff is one test-support file. skip-changeset is declared with this PR.

PR and merge-group CI never run the dogfood legs, so this PR's checks cannot show the fix. The card's acceptance is read on main after landing: the first push run that runs Dogfood Regression Gate (3/3), and the next scheduled run. Both run ids will be reported on the card.

Acceptance notes

• A docblock that links the handler by name moves this control, and so do the nine sibling domain rows, each of which counts its own handler's name by substring. That is the pin doing its job: it makes whoever touches the file re-read the row. It is also why PR CI, which skips the dogfood legs, merged the change green. Making PR or merge-group CI run them is out of scope here. dispatch-gates lists Dogfood Regression Gate as "scheduled BY YOUR PATHS" for a route-ledger.ts change, but its if: skips pull_request and merge_group, so neither PR CI nor the derived commands run the dogfood pins (main red after #24009) #24073 and [Decision] The required Test Core and Dogfood Regression Gate go red on superseded runs (139 in 51 h, 0 real catches): keep #16157's ruled reading, or decide by the cancel cause the platform records? #22653 ruled on that.

---

Generated by Claude Code

[github-actions]

📓 Docs Drift Check

Nothing in this diff resolved to a documentable surface (no symbol, route or SDK anchor derived from 0 changed package(s)), so this run has no opinion about the docs.

What this run could not see

• a page that states a rule by its inputs shares no identifier with the emitter that implements the rule, so an emitter-only diff cannot list it — not on this run and not on any run. Measured on fix(driver-sql): emit varchar(maxLength) for a text field a declared index keys on #11430: content/docs/protocol/objectql/types.mdx documents the text-family column mapping by the ObjectQL type names it maps FROM (text / textarea / html) while the diff changed createColumn; it went unlisted, and it was the page that diff falsified, in four places. No shared token exists to detect this on, so a rule your change carries has to be re-read by hand in the pages that restate it.
• a key NAME is not a key, so the hand re-read the line above prescribes can land on the wrong schema. The same spelling is authorable on one governed type and a [REMOVED] tombstone on another for each of active, aria, joins, objects, template, tools and version (censused on [finding] tools is a key on BOTH AgentSchema (tombstoned, dead) and SkillSchema (live, cloud-attested), so a name-based search attributes skill examples to the agent key — it produced a false stop-the-line alarm on PR #19059 #19093 over the liveness ledger's governed types, top-level keys); nothing in a search result distinguishes the two, so a grep hit on a LIVE example reads as evidence about the DEAD key. Measured on fix(spec): the agent.tools liveness row says dead — it claimed live on a key the schema tombstoned #19059: content/docs/ai/agents.mdx was reported as contradicting the agent.tools tombstone over its tools: example at :161, which is inside the defineSkill({ block opened at :155 — the page was already correct. Settle ownership by PARSING the value against both schemas, never by the name: that literal PASSES SkillSchema, and as an AgentSchema it FAILS at tools with the tombstone prescription. ⛔ These names are not the whole class — a key retired through a .strict() guidance map leaves no tombstone in the walked shape and none of them here (tool.category, live as AIToolDefinition.category).

Coarse fallback — 0 page(s) merely mention a changed package (the pre-#9192 predicate, kept for the deliberately-wide backstop): node scripts/docs-audit/affected-docs.mjs --json a4eae06499a8f0377960bfc9c98daced5cf19258 → packageMentionDocs.

[objectstack-fleet]

席位核对

Verdict: ACCEPT at head ed1c5b39608d95086e868a16e5b09c51d12eb2df. The PR lands once CI is green, under the standing grant. It is not Tier H.

Checklist

• PR shape: draft → main, Fixes #25988. One file, packages/qa/dogfood/test/authz-probe-blind-spot.census.ts (+12/−4). check-governed-merges --pr 26002: NOT governed, 16 lines. skip-changeset is correct, since this is a private test census and nothing ships.
• Contract review: not triggered (Clause-②: no, no packages/spec/src/**, no governed text).

The pin was re-read, not just bumped

• I checked this against fix(runtime): GET /i18n/translations/:locale carries an ETag and Cache-Control: private, no-cache, and a matching If-None-Match is answered 304 (#25785) #25956 itself (git show 69fa23d6 -- packages/runtime/src/domains/i18n.ts). The only added line naming the handler is the docblock {@link handleI18nRequest} in TRANSLATIONS_CACHE_CONTROL's comment, at origin/main i18n.ts:73. The functions it added are module-private requestHeader, ifNoneMatchHits and translationsAnswer. It adds no export async function handle*Request, no DomainRoute and no shouldDenyAnonymous(. So the row's population, reach and gate count stay at 1/1/1, and only the control count moves, from 3 to 4. The row comment records why.

Seat's own reading, in a detached worktree at ed1c5b3

• npx vitest run test/authz-probe-blind-spot.test.ts in packages/qa/dogfood → Tests 47 passed (47). On origin/main cdd4eea938 the same command gave 1 failed | 46 passed, with the same assertion as dogfood: main's Dogfood Regression Gate (3/3) is red since #25956 — the authz blind-spot census still pins handleI18nRequest at 3 in packages/runtime/src/domains/i18n.ts; #25956 made it 4 #25988.

Gates

• CI was still converging when I read it (16 success, 16 skipped, 18 in progress, 0 failed). Before landing I check Lint & Repo Gates and TypeScript Type Check on this head.
• PR and merge-group CI do not run the dogfood legs (dispatch-gates lists Dogfood Regression Gate as "scheduled BY YOUR PATHS" for a route-ledger.ts change, but its if: skips pull_request and merge_group, so neither PR CI nor the derived commands run the dogfood pins (main red after #24009) #24073). The card's acceptance is read after landing: the first main push run that runs Dogfood Regression Gate (3/3), and the next scheduled run.

---

Generated by Claude Code