test(dogfood): the two view resets drop the ?package= the reset door never read

[objectstack-fleet]

Fixes #23084
Clause-②: no

What changes

packages/qa/dogfood/test/view-container-cross-package-default.dogfood.test.ts, two lines: the two view resets drop the ?package=${REPAIR} the reset door never read.

• :174 DELETE /meta/view/os_qa_shadow_probe?package=${REPAIR} becomes DELETE /meta/view/os_qa_shadow_probe
• :228 DELETE /meta/view/os_qa_keyed_probe?package=${REPAIR} becomes DELETE /meta/view/os_qa_keyed_probe

Nothing else moves. The two PUT saves (:149, :203) keep their ?package=, because the save door reads it. The closed query set in packages/rest/src/rest-server.ts is not widened. No changeset: @objectstack/dogfood is private and publishes nothing.

Why (triage ruling 6131531682, applied as ruled)

• The hourly full run has been red on Dogfood Regression Gate (1/3) since 038d4c9588 (PR feat(metadata-protocol): one language of an email template is addressed by name — the by-name /meta doors accept a bundle member selector (#22877, stage 2) #23070). That PR gave DELETE {metaPath}/:type/:name its closed query set, META_ITEM_RESET_PARAMS = ['state', 'dropStorage', member] (rest-server.ts:744, refused at :7346), as route rule 5 asks. package is not in it, so these two test calls now get a correct located 400.
• The reset handler never read package. On 2d21b9c891 the handler's only query reads are state (:7348), dropStorage (:7357) and the member selector (:7386). The four helpers it calls (resolveExecCtx, walledCatalogRefusal, resolveProtocol, resolveMetaWriteActor) read no req.query at all. DeleteMetaItemRequest declares no package (protocol.lock-one-resolution.test.ts:607-608 says so). The pre-feat(metadata-protocol): one language of an email template is addressed by name — the by-name /meta doors accept a bundle member selector (#22877, stage 2) #23070 handler (038d4c9588^) read the same two names, state and dropStorage, and nothing else. So dropping the parameter changes no behaviour, because the handler already ignored it.
• git grep on 2d21b9c891: these two test lines are the only places that put ?package= on a meta DELETE.

Measured

All runs are in this container, on base 2d21b9c891 plus the fix a61ad11dd3, through scripts/pm/os-verify-lock.sh. The dogfood dependency closure was built first (pnpm turbo run build --filter=@objectstack/dogfood, 63/63 tasks).

• Reproduction before (unmodified file at 2d21b9c891): Tests 2 failed | 2 passed (4). Both failures are AssertionError: expected 400 to be 200, at :175:28 (steps 1-7) and :229:28 (keyed member). This matches the hourly run.
• After (a61ad11dd3): the same file gives Test Files 1 passed (1), Tests 4 passed (4), and the lock reports VERDICT command-exit 0.
• Shard 1/3 (OS_TEST_SHARD=1/3 pnpm --filter @objectstack/dogfood exec vitest run --maxWorkers=2, CI's own carrier, at a61ad11dd3): Test Files 87 passed (87), Tests 718 passed (718), and the lock reports VERDICT command-exit 0.
  • The 87 files match vitest 4.1.11's BaseSequencer.shard slice (sha1 of the root-relative path, [0, 87) of 259), recomputed locally. That slice contains this file.
• Typecheck: pnpm --filter @objectstack/dogfood run typecheck exits 0. tsc --listFiles counts the edited file once, so the program does cover it.
• Derived gates: dispatch-gates --commands lists 53 commands, the same list on 2d21b9c891, on a61ad11dd3 with no paths, and on current origin/main abf86554c9. All 53 exit 0 at a61ad11dd3.
  • check:dual-build-cjs-loads first exited 3 (PREREQUISITE NOT MET, 8 packages unbuilt). It exited 0 once those were built: 107 published require entry point(s) across 66 package(s) load.
  • --ran reconciliation: 53 derived famil(ies) accounted for.
• Lint, proven narrowing over the one touched file:
  1. Population, read from eslint's own config: isPathIgnored: false, with 5 rules enabled for the file.
  2. File count, from --format json: 1 file, 0 errors, 0 warnings, exit 0.
  3. Invariance: the resolved config has no parserOptions.project or projectService, and the 5 rules are per-file AST rules. So this diff cannot move any other file's verdict.

For #23072: the PUT door does read package

PUT {metaPath}/:type/:name reads package at rest-server.ts:7175 (metaItemPackageBinding(req.query?.package)) and threads it as packageId into the save request (:7225).

The handler's other query reads:

• force (:7166);
• mode, read a frame down in metaSaveRequestOptions (:7146, meta-save-request.ts:94);
• the repeat guard names ['force', 'package', 'mode'] (:7145).

The PUT declares no closed set yet. The test's PUT ...?package=${REPAIR} at :203 carries weight: :222 asserts _packageId is REPAIR. This line is not changed here.

Acceptance notes

• Not merged with origin/main: it moved 4 commits during this run (abf86554c9: lint, metadata-protocol, objectql, CI timing scripts). This diff is two test lines and dogfood runs only in the full battery, so the pin is the next hourly full run on main: Dogfood Regression Gate (1/3) green.
• Shards 2/3 and 3/3 were not run locally, by declared narrowing. The edited file is in shard 1, and every other dogfood file boots its own app or a per-worker shared one that this file is not part of. CI's full battery runs all three.
• The same hourly run's Test Core (6/6) red is shard-timing-drift and is not this card, as the claim notes.

---

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 55828c5145ce1ca8cb597baf3396614adf9c0583 → packageMentionDocs.