[objectstack-fleet]

Requested by Leehom · project thread

Fixes #26049

Clause-②: no
Metadata-protocol: none

What timed out, read from the logs

Both merge-queue ejections (build 40300684735 for #26012, Test Core 8/8; build 40302541664 for #26007, Test Core 8/8) fail at src/hook-run-as.test.ts:389:5, the case "the seed is no nameable member of the published face: keyed on a symbol the package entry does not export". The triage comments truncate the FAIL line after the describe title, which made it read as the block's first case; the full title and the line number are in the job logs' ::error annotation.

That case did const entry = await import('./index.js') inside the test body. The package entry pulls the rest of @objectstack/objectql's module graph, which nothing else in this file loads, so the case paid a cold transform and import of the package inside its 5000 ms budget. The resolver path the card suspected is not the cost: resolveUserAuthzGrants is a static import in engine.ts, and the block's first case runs in 1 ms.

The fix (test-only)

import * as entry from './index.js' at the top of the file, the same idiom the package's *-code-constant.test.ts files use for the barrel. The load now happens at file collection, outside any per-case window. The case becomes synchronous and keeps all five assertions unchanged (SEED_USER_GRANTS and APP_AUTHORED_HOOK_API absent from the entry, withUserGrantsSeeded absent from the prototype, two symbol-keyed members). No quarantine, skip, retry or timeout change. No product code touched.

Evidence

Run: pnpm --filter @objectstack/objectql exec vitest run --maxWorkers=2 --reporter=verbose src/hook-run-as.test.ts, under scripts/pm/os-verify-lock.sh, with 8 node -e 'for(;;){}' busy loops alongside on a 4-core container (2x oversubscribed, standing in for the ui lanes running beside Test Core). Shared-box seconds.

	case time	file result	tests phase
quiet box, origin/main 9753d16	2513 ms	29/29	2.61 s
under load, origin/main, 5 runs	5016, 5021, 5366, 5413, 5070 ms (all Test timed out in 5000ms.)	5 of 5 red, 1 failed / 28 passed	5.27–5.73 s
under load, this branch, 5 runs	1, 0, 0, 0, 1 ms	5 of 5 green, 29/29	0.20–0.37 s

The load did not disappear; it moved. Under load the file's import phase went from 12.8–17.9 s (before) to 20.3–26.3 s (after), and whole-file wall time stayed in the same 18–27 s band. That phase is not bounded by testTimeout.

Local verification at the final head: pnpm --filter @objectstack/objectql typecheck exit 0 (check:test-typecheck OK, test-layer debt held); eslint packages/objectql --no-inline-config exit 0; pnpm --filter @objectstack/objectql test exit 0 (470 files / 8748 tests passed); the derived gate list (dispatch-gates.mjs --commands) 58 derived, 57 run with exit 0, 1 NOT MEASURED (check:dual-build-cjs-loads, exit 3: needs a full build); --ran reconciles 58/58 at d8e4f756c.

Changeset

skip-changeset: the diff is one *.test.ts file, and @objectstack/objectql's files[] is dist, README.md, CHANGELOG.md, so nothing published changes.

Acceptance notes

• packages/objectql/src/bound-action-collision-report.test.ts:111 has the same shape (await import('./index.js') inside a "no nameable member of the published face" case). It has not ejected a queue build, and it is outside this claim's file surface, so it is untouched here; reported to the seat.

🤖 Generated with Claude Code

https://claude.ai/code/session_01AN1CuLrAYkcsfrh1N5ffLT

---

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 1bbb7bf4908f634a5b3164305062d117d4b8b4fe → packageMentionDocs.

[objectstack-fleet]

席位核对 · ACCEPT · domain:server seat 1 · os-project-manager · session_01R1q2xdtYgU6FKrZ7LgjPCC · 2026-10-26T13:41Z

Head d8e4f756c0, report 6416939033 (on #26049), claim 6415849022. No contract review owed: the change is test-only, with no packages/spec/** diff and no narrowing.

item	reading
Shape	draft, base main, Fixes #26049, skip-changeset (test-only), mergeable clean.
Files	1 file, +7/−2: packages/objectql/src/hook-run-as.test.ts.
Root cause	Both ejections were the line-389 case (the seed is no nameable member of the published face), not the block's first case as the order hypothesised; the triage FAIL line was truncated. That case did await import('./index.js') inside its 5 s budget. The entry pulls the package's whole module graph, about 2.5 s on a quiet box, and the queue's load pushed it past 5 s.
Fix	The entry is imported statically with the file, so the load is paid at collection, outside any case budget. Same module, same assertions; no timeout raised, no skip. Under 8 CPU hogs: 5/5 red on main (5016–5413 ms), 5/5 green after (0–1 ms).
Evidence	36 check runs success and 16 expected skips on this head, 0 failed. Build Core (which hosts check:dual-build-cjs-loads, the one gate the dev could not measure locally) is green. objectql suite 470 files / 8,748 tests green, typecheck and scoped lint green.
Carried	The dev noted bound-action-collision-report.test.ts:111 has the same in-case import() shape. It has not ejected a PR, so no card; the anchor workflow files one if it does.
Landing	「按照 pm-dispatch 规则复核过的都授权」 (Leehom, 2026-10-25T11:35Z): ready and auto-merge through the relay.

---

Generated by Claude Code