ci: repo-scanning tests move out of the heavy packages' test tasks into one repo-conformance task so package suites become cacheable again (maintainer-directed, measure-first)

[claude]

Filed by the skills lane seat (session session_019RfFHiRCSs3JXLK4cwcfox, os-steve) on the maintainer's direction, 2026-09-07T02:5xZ, and dispatched by the same seat on that direction. Surface owner stays domain:devx.

Authority (maintainer, verbatim, live PM chat 2026-09-07): 「同意你的建议，你负责执行派发所有可行的优化」, after the seat measured why docs-only merges still run 15–20-minute shards in the queue.

What

The repo-scanning tests — the ones whose declared inputs are the whole repository — leave the heavy packages' test tasks and run once per CI run as their own task, so @objectstack/spec, core, types, runtime, objectql and rest become cacheable again and a change outside a package no longer re-runs that package's whole suite.

Measured

• turbo.json declares $TURBO_ROOT$ inputs 169 times. @objectstack/spec#test lists packages/**/*.ts, content/**, skills/**, scripts/** and more; core#test, types#test, runtime#test, objectql#test list packages/**/*.ts; rest#test lists content/**, docs/**, skills/**. Any .ts or docs change anywhere invalidates those five or six suites — the largest in the repo (spec alone 11,420 cases).
• The declarations are correct: scripts/cross-package-test-inputs.mjs is the registry of tests that read outside their package (117 rows), and check-cross-package-test-inputs.mjs keeps turbo's inputs equal to it (main is red for every PR that touches packages/spec: #7769 gave sys_api_key update without bulk, and the conformance scan that catches it lives in a package #7769 never touched #7802). The cost is structural, not a mistake: package-scoped caching cannot work while repo-scanning tests live inside package suites.
• Measured consequence: docs-only merge groups at 01:18Z ran shards of 15–20 minutes.

Ruling — measure first

1. From the registry, list every repo-scanning test file with its home package and the globs it reads; measure each file's duration (a local vitest run of those files, or the per-file report from ci: every Test Core run publishes the slowest test files and packages beside their pinned weights (maintainer-directed, part B measurement) #16454 once it exists) and the share of its package's suite time. Report the table before any edit.
2. If the scanning files are at most ~60 and separable (they import from their package's public entry or from the repo, not from private internals), move them into one task — a repo-conformance vitest project (home to be measured: packages/qa or a root-level project) that owns the wide inputs and runs once per CI run as its own Test Core item; the packages' test tasks then keep package-local inputs plus their dependency graph, and the cross-package registry keeps its gate over the new home. If the files are more numerous or entangled with package internals, stop after the table and report options with a recommendation on the four axes.
3. Nothing about what the tests assert changes; no test is deleted; the registry gate stays green under the new layout.
4. Clause-②: yes — what CI verifies per package at merge time changes shape. Contract tier.

Acceptance

• The table from step 1 in the PR body; turbo cache hit rates before/after for a docs-only change and a leaf change, measured locally (turbo run test --dry-run=json shows the cache status per task).
• check-cross-package-test-inputs green; partition-test-shards --self-test green with the new item present.

Refs #16173, #16453.

---

Generated by Claude Code

[claude]

This lane's execution seat is standing off this card — half-state recorded, not claimed (PM seat domain:devx @ objectstack (#6023), session session_01Vbw3RPgdtqesx4azk9SbW8, read 2026-09-07T03:30Z).

The body says this card was "dispatched by the same seat on that direction" (skills lane, session_019RfFHiRCSs3JXLK4cwcfox, os-steve), and that seat's neighbouring dispatches in this lane are visible and consistent with it: #16453 is pm:dispatched assigned to os-steve (PR #16471), and #16445 carries that seat's Claim: from 01:45Z (PR #16451). On GitHub, though, this card carries no assignee, no Claim: comment and no branch — so the dispatch it declares is not readable by the protocol that decides ownership, and a second seat reading only the labels would take it.

This seat is not claiming it: a card whose body names another session's dispatch is that session's to finish, and the maintainer's direction quoted above routes it there. Recording the gap instead, so the card stops looking free:

• If os-steve's dev is flying: post the Claim: (session + branch) and set the assignee, and this note needs nothing further.
• If no Claim: appears and the card is still bare, the dispatch did not survive; then it is an ordinary pm:queue card in this lane and this seat will claim it on a later check-in. Restart condition for this seat: pm:queue, no assignee, no Claim:, no issue-16466 branch, and no PR referencing it.

No labels, assignee or title were written by this note. Same reading applies to #16469 only in reverse — that one carries no such declaration, so this seat is claiming it now.

---

Generated by Claude Code

[claude]

Claim: PM loop round 9 — flight C3, maintainer-directed (「同意你的建议，你负责执行派发所有可行的优化」, 2026-09-07): the repo-scanning tests leave the heavy packages' test tasks, measure-first — the registry (scripts/cross-package-test-inputs.mjs, 117 rows) enumerated with each file's duration and import style before any edit; at most ~60 separable files ⇒ one repo-conformance vitest project owns them and the wide $TURBO_ROOT$ inputs, the packages' test tasks keep package-local inputs, the registry gate stays green; more or entangled ⇒ stop after the table with options on the four axes. No assertion changes, no test deleted.
Session: session_019RfFHiRCSs3JXLK4cwcfox
Branch: claude/issue-16466-repo-conformance-task
Worktree: objectstack-issue-16466
Domain: domain:devx (dispatched by the skills seat on the maintainer's direction)
File surface: turbo.json (the test tasks' inputs), scripts/cross-package-test-inputs.mjs (rows re-homed, never deleted), the moved test files and the new project's config, one timing row or the fallback for the new item. ⛔ ci.yml untouched; scripts/test-shard-timings.json written only through the documented generator.
Container & model: L, mode:subagent, fable — Clause-② yes (what CI verifies per package at merge time changes shape); on a fable 429 the seat's exemption applies, opus, recorded here.
Clause-②: yes
Serial: the timings JSON is C2a's (#16464, in flight) — merge-tree prove clean against its pushed head twice before pushing, or take the fallback row and say so; turbo.json has no open PR. Not governed unless a governed path is touched ⇒ in-seat contract-tier review, then the seat flips ready and arms SQUASH auto-merge; changeset per the gate's rule. The PR carries Fixes #16466.

---

Generated by Claude Code

[claude]

#16466 measurement table (ruling step 1) — read before the report comment below

Tree measured: a9d25d842f (worktree base; origin/main had moved to 4500bc5d79 by the end of the run — of the files this card measures only .github/workflows/ci.yml changed in between, PR #16471's merge_group affected-set selection; the registry, turbo.json, the timings dataset, partition-test-shards.mjs and every listed test file are byte-identical across that range).

Population: node scripts/check-cross-package-test-inputs.mjs --list-escapes — 27 packages, 113 escaping test files; the registry (scripts/cross-package-test-inputs.mjs) holds 168 glob rows across those 27 entries. The card's "117 rows" is the gate's self-test case count (All 117 self-test cases passed), not the registry's size.

Method. Duration = per test module, collectDuration + setupDuration + prepareDuration + environmentSetupDuration + duration from vitest 4.1.10's TestModule.diagnostic() through a 20-line custom reporter (the stock json reporter reports only the run phase, and these files do their repo scans at module top — types' one escaping file is 5.5s of collect and 0.13s of run). "Isolated" = one vitest run per package over its escaping files only (--maxWorkers=2); "in-suite" = the same reporter over the package's full local suite, measured for the six heavy packages (spec in three --shard legs). Every run went through scripts/pm/os-verify-lock.sh (slot issue-16466), so absolute seconds are shared-box readings — sibling agents ran unlocked builds and gates on the same cores; ratios and shares are the durable numbers. Package suite denominators: local full-suite sum for the six heavy packages; the CI-median from scripts/test-shard-timings.json (2026-08-24) for the rest, labelled as a mixed-machine estimate. @objectstack/dogfood's 9 files are listed but not measured (excluded from Test Core by --exclude @objectstack/dogfood, run by its own file-sharded job, and each boots a real app).

Import style = every relative import resolved against the file's own package directory: public entry (bare @objectstack/* only), fs-only (no module import of its subject at all), PRIVATE internals (a relative import inside its own package), repo path (a relative import that leaves the package, e.g. scripts/js-comment-mask.mjs).

Per-package summary (sorted by CI-median suite wall)

package	escaping files	measured	sum s (isolated)	package suite (CI median s)	wide globs held	import styles
@objectstack/cli	27	27	244.1	458.15	—	private+escape=11, private-internals=13, relative-escape=3
@objectstack/spec	26	26	58.8	403.65	content/, packages//.object.ts, packages/**/.ts, scripts/, skills/	fs-only=9, private+escape=1, private-internals=15, relative-escape=1
@objectstack/metadata-protocol	1	1	6.7	278.86	—	private-internals=1
@objectstack/runtime	2	2	20.9	264.23	packages/**/*.ts	private+escape=1, private-internals=1
@objectstack/rest	1	1	0.4	236.68	content/, docs/, skills/**	private-internals=1
@objectstack/objectql	1	1	0.4	222.2	packages/**/*.ts	private-internals=1
@objectstack/plugin-auth	3	3	2.1	221.71	packages/**/*.object.ts	private-internals=1, public-entry=1, relative-escape=1
@objectstack/driver-sql	1	1	8.8	203.97	—	private+escape=1
@objectstack/plugin-security	1	1	2.0	114.92	—	private-internals=1
@objectstack/lint	7	7	22.7	86.66	—	private+escape=3, private-internals=4
@objectstack/metadata	1	1	0.3	43.03	—	private+escape=1
@objectstack/service-settings	1	1	0.4	41.61	—	private+escape=1
@objectstack/plugin-email	1	1	7.0	39.1	—	private+escape=1
@objectstack/mcp	1	1	8.7	34.44	—	private+escape=1
@objectstack/cloud-connection	2	2	8.8	33.64	—	private+escape=2
@objectstack/client	7	7	20.8	32.29	scripts/**	private+escape=5, relative-escape=2
@objectstack/plugin-approvals	2	2	21.2	31.07	—	private+escape=2
@objectstack/trigger-record-change	1	1	14.7	23.53	—	private+escape=1
@objectstack/example-showcase	1	1	11.6	22.87	—	private+escape=1
@objectstack/platform-objects	2	2	19.7	18.78	packages/**/*.object.ts	private+escape=1, public-entry=1
create-objectstack	8	8	23.8	15.5	content/**	fs-only=4, private-internals=4
@objectstack/core	2	2	4.9	15.1	packages/**/*.ts	private-internals=1, public-entry=1
@objectstack/formula	2	2	7.2	12.39	—	private-internals=2
@objectstack/downstream-contract	1	1	3.7	9.88	packages/spec/src/**	public-entry=1
@objectstack/types	1	1	1.9	5.69	packages/**/*.ts	fs-only=1
@objectstack/trigger-api	1	1	0.4	1.72	—	private+escape=1
@objectstack/dogfood	9	0	0.0	—	examples/app-showcase/**	private-internals=6, public-entry=2, relative-escape=1

The six heavy packages — local full suites and the escaping share

package	suite files	local full-suite sum s	local wall s	escaping files	escaping in-suite s	escaping share
@objectstack/spec	482	736.7	527.7	26	53.9	7.3%
@objectstack/core	51	59.7	43.2	2	2.4	4.0%
@objectstack/types	20	17.7	13.0	1	5.7	32.1%
@objectstack/runtime	238	570.1	307.7	2	5.6	1.0%
@objectstack/objectql	284	531.1	290.9	1	0.4	0.1%
@objectstack/rest	187	337.9	186.8	1	0.4	0.1%

All six suites green locally (spec 4544+4118+4440, runtime 3361, objectql 4800, rest 3179, core 1232 with one contention timeout on the second pass — filed as #16497 — types 578).

The decision-bearing rows: who holds each wide glob, and how that file imports its subject

package	wide glob(s)	holder file	imports its subject via	bindings reachable through the package's public entry (built dist/*.d.ts)?
@objectstack/spec	packages/**/*.ts, content/**, scripts/**, skills/**, examples/** (all heldBy-witnessed)	src/system/compliance-families-retirement.test.ts — walks FIVE repo roots by readdirSync on a loop variable	PRIVATE: ../../scripts/lib/export-origins-testkit (EXPORT_ENTRY_POINTS, exportNamesOf, holdersOf), ../migrations/registry, dynamic ./index, ./incident-response.zod	no — the testkit is a private script lib (0 hits in any dist/*.d.ts); the migrations registry is public
@objectstack/spec	packages/**/*.ts (also)	scripts/liveness/evidence.test.ts · scripts/publish-smoke-port-collision.test.ts · src/shared/retired-key-migrate-sentence.test.ts · src/ui/action-requires-confirmation-docblock.pin.test.ts	PRIVATE ./evidence.mts · fs-only · fs-only · fs-only	— (moving these four drops nothing while the compliance pin stays: its walk covers the same radius)
@objectstack/spec	content/** (also)	scripts/file-description.test.ts · scripts/query-pointer-row.test.ts · scripts/root-index.test.ts · src/api/error-catalog-docs.test.ts	PRIVATE ./lib/file-description (+8 more) · PRIVATE ./lib/file-description · PRIVATE ./lib/sharded-artifacts, ./lib/root-index · PRIVATE ./errors.zod	no (scripts/lib/* are unpublished)
@objectstack/spec	skills/** (also)	scripts/export-list.test.ts · src/shared/retired-key-migrate-sentence.test.ts	PRIVATE ./lib/export-list · fs-only	no / n.a.
@objectstack/core	packages/**/*.ts	src/security/authz-store-unavailable.test.ts · src/security/operation-private-keys.pin.test.ts	PRIVATE ./resolve-authz-context.js, ./authz-store-unavailable.js · public @objectstack/core	yes — all 7 bindings are in dist/index.d.ts · already public
@objectstack/types	packages/**/*.ts	src/driver-error-classification.callers.test.ts	fs-only (+ typescript)	n.a. — separable as is
@objectstack/runtime	packages/**/*.ts	src/action-owner-key-single-source.test.ts	PRIVATE ./action-execution.js (standaloneActionObjectName, standaloneActionOwnerKey re-export, type ActionExecutionDeps) + public @objectstack/objectql	no — the two value bindings have 0 hits in dist/*.d.ts; the pin exists to check that private re-export
@objectstack/runtime	(narrow only: scripts/js-comment-mask.*)	src/error-envelope.conformance.test.ts	PRIVATE ./http-dispatcher.js, ./error-envelope.js, ./dispatcher-error-vocabulary.js + repo path scripts/js-comment-mask.mjs	holds no wide glob — need not move
@objectstack/objectql	packages/**/*.ts	src/action-owner-key-single-source.test.ts	PRIVATE ./action-governance.js (GLOBAL_ACTION_OBJECT_KEY, standaloneActionOwnerKey)	yes — both in dist/index.d.ts
@objectstack/rest	content/**, docs/**, skills/**	src/meta-state-route-doc-spelling.test.ts	PRIVATE ./rest-route-ledger.js (REST_ROUTE_LEDGER)	no — 0 hits in dist/*.d.ts (packages/client already reaches it by a relative cross-package path, declared in the registry)

Count for the ruling's step 2, heavy six: 33 files; 20 import PRIVATE internals of their home package; 11 fs-only; 1 public entry; 1 repo path. Separable as they stand (no private-internal import, or every private binding provably re-exported from the package's public entry): core 2, types 1, objectql 1 (four files), plus spec 10 (nine fs-only and schema-tree-freshness.test.ts) that hold no wide glob the compliance pin does not also hold. Entangled: spec 16, runtime 2, rest 1.

"Before" readings (card acceptance; no "after" exists because no edit was made)

turbo run test --dry-run=json on the clean tree, then under a scratch docs-only edit (content/docs/concepts/north-star.mdx) and a scratch leaf edit (packages/connectors/connector-slack/src/index.ts, whose only dependent is examples/app-showcase), each restored with git checkout HEAD -- and proven by git hash-object equal to the HEAD blob and an empty git diff HEAD. 147 tasks, 79 of them #test. The local cache holds no test entries (every test here ran through vitest, never through turbo), so cache.status reads MISS on both sides of each edit; the reading that does not depend on a warm cache is whose task hash moved, which is the same thing on a warm cache: an unmoved hash is a hit.

• docs-only edit → 3 #test hashes move: @objectstack/spec#test, @objectstack/rest#test, create-objectstack#test. check-cross-package-test-inputs.mjs --union-into from an EMPTY affected set unions the same three (+ @objectstack/spec, + @objectstack/rest, + create-objectstack). With ci: the merge queue's Test Core runs the affected package set; selection moves into a script with a self-test #16471 landed, that is the whole Test Core package set of a docs-only merge group: CI-median 403.65 + 236.68 + 15.5 = ~656s of suite wall for a group with nothing to test, of which the escaping files themselves are 53.9 + 0.4 + ~24s.
• leaf .ts edit → 8 #test hashes move: the three genuinely affected (connector-slack, example-showcase, dogfood) plus spec, core, types, runtime, objectql; --union-into from an empty set unions exactly those five (CI-median ~911s). rest#test does not move.

The full 113-row table

† = held by heldBy witness (a read the detector cannot name: loop variable or directory listing). Bold globs are the wide ones this card is about.

#	home package	test file	globs it holds (declared radius)	imports its subject via	duration s (collect+run, isolated)	share of package suite
1	@objectstack/cli	packages/cli/src/commands/generate-builtin-id-column.pin.test.ts	packages/drivers/driver-sql/src/sql-driver.ts	PRIVATE internals: ./generate.js	1.82	~0.4% of CI-median 458s (mixed-machine estimate)
2	@objectstack/cli	packages/cli/src/commands/generate-field-type-vocabulary.pin.test.ts	packages/spec/src/data/field.zod.ts, packages/drivers/driver-sql/src/schema-drift.ts	PRIVATE internals: ./generate.js	3.74	~0.8% of CI-median 458s (mixed-machine estimate)
3	@objectstack/cli	packages/cli/src/commands/generate-multiple-json-column.pin.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./generate.js	1.29	~0.3% of CI-median 458s (mixed-machine estimate)
4	@objectstack/cli	packages/cli/src/commands/generate-string-family-width.pin.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./generate.js	15.02	~3.3% of CI-median 458s (mixed-machine estimate)
5	@objectstack/cli	packages/cli/src/commands/serve-audit-registration.contract.test.ts	scripts/js-comment-mask.mjs	repo path: scripts/js-comment-mask.mjs	0.13	~0.0% of CI-median 458s (mixed-machine estimate)
6	@objectstack/cli	packages/cli/src/commands/serve-bound-port-publication.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./serve.js, ../utils/port-contract.js + repo path: scripts/js-comment-mask.mjs	12.40	~2.7% of CI-median 458s (mixed-machine estimate)
7	@objectstack/cli	packages/cli/src/commands/serve-cluster-host-resolution.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ../utils/tenancy-posture-hints.js, ../utils/spelling.js, ./deeper.js, ../utils/deeper.js, ../utils/does-not-exist.js + repo path: packages/other-pkg/spelling.js	0.29	~0.1% of CI-median 458s (mixed-machine estimate)
8	@objectstack/cli	packages/cli/src/commands/serve-email-config-parity.contract.test.ts	scripts/js-comment-mask.d.mts	PRIVATE internals: ./serve.js + repo path: scripts/js-comment-mask.mjs	4.60	~1.0% of CI-median 458s (mixed-machine estimate)
9	@objectstack/cli	packages/cli/src/commands/serve-multi-node-cap-advisory.pin.test.ts	packages/services/service-cluster/src/**	repo path: scripts/js-comment-mask.mjs	0.22	~0.0% of CI-median 458s (mixed-machine estimate)
10	@objectstack/cli	packages/cli/src/commands/serve-verify-security-parity.contract.test.ts	packages/verify/src/, packages/plugins/plugin-security/src/	repo path: scripts/js-comment-mask.mjs + @objectstack/plugin-security	2.68	~0.6% of CI-median 458s (mixed-machine estimate)
11	@objectstack/cli	packages/cli/src/utils/console-route-ledger.conformance.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./console-route-ledger.js + repo path: scripts/js-comment-mask.mjs	0.99	~0.2% of CI-median 458s (mixed-machine estimate)
12	@objectstack/cli	packages/cli/src/utils/port-contract-single-source.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./port-contract.js, ../utils/port-contract.js + repo path: scripts/js-comment-mask.mjs	0.63	~0.1% of CI-median 458s (mixed-machine estimate)
13	@objectstack/cli	packages/cli/test/cloud-login-json-ndjson.e2e.test.ts	content/docs/deployment/cli.mdx, content/docs/deployment/index.mdx	PRIVATE internals: ./helpers/serve-process.js	29.01	~6.3% of CI-median 458s (mixed-machine estimate)
14	@objectstack/cli	packages/cli/test/create-plugin-docs-parity.test.ts	content/docs/plugins/index.mdx, content/docs/protocol/kernel/index.mdx, content/docs/protocol/kernel/plugin-spec.mdx	PRIVATE internals: ../src/commands/create.js	2.65	~0.6% of CI-median 458s (mixed-machine estimate)
15	@objectstack/cli	packages/cli/test/i18n-section-coverage.test.ts	packages/spec/src/system/translation.zod.ts, examples/app-showcase/src/ui/views/contact.view.ts, examples/app-showcase/src/data/objects/semantic-zoo.object.ts	PRIVATE internals: ../src/utils/i18n-extract, ../src/utils/i18n-coverage, ../src/commands/lint + repo path: examples/app-showcase/src/ui/views/contact.view, examples/app-showcase/src/data/objects/semantic-zoo.object	11.79	~2.6% of CI-median 458s (mixed-machine estimate)
16	@objectstack/cli	packages/cli/test/i18n-tab-coverage.test.ts	examples/app-showcase/src/ui/pages/task-triage.page.ts	PRIVATE internals: ../src/utils/i18n-extract, ../src/utils/i18n-coverage + repo path: examples/app-showcase/src/ui/pages/task-triage.page	2.03	~0.4% of CI-median 458s (mixed-machine estimate)
17	@objectstack/cli	packages/cli/test/init-created-files-summary.e2e.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./helpers/serve-process.js	23.35	~5.1% of CI-median 458s (mixed-machine estimate)
18	@objectstack/cli	packages/cli/test/init-template-comments-self-contained.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ../src/commands/init.js, ../src/commands/create.js	2.55	~0.6% of CI-median 458s (mixed-machine estimate)
19	@objectstack/cli	packages/cli/test/login-json-ndjson.e2e.test.ts	content/docs/permissions/authentication.mdx	PRIVATE internals: ./helpers/serve-process.js	18.97	~4.1% of CI-median 458s (mixed-machine estimate)
20	@objectstack/cli	packages/cli/test/login-json-noninteractive.e2e.test.ts	scripts/check-nul-bytes.mjs	PRIVATE internals: ./helpers/serve-process.js	71.08	~15.5% of CI-median 458s (mixed-machine estimate)
21	@objectstack/cli	packages/cli/test/published-entry-stderr-nonblocking.e2e.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./helpers/serve-process.js + repo path: scripts/js-comment-mask.mjs	0.27	~0.1% of CI-median 458s (mixed-machine estimate)
22	@objectstack/cli	packages/cli/test/run-dev-stderr-nonblocking.e2e.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./helpers/serve-process.js + repo path: scripts/js-comment-mask.mjs	2.59	~0.6% of CI-median 458s (mixed-machine estimate)
23	@objectstack/cli	packages/cli/test/scaffold-emission-policy.e2e.test.ts	content/docs/getting-started/index.mdx, content/docs/deployment/troubleshooting.mdx	PRIVATE internals: ./helpers/serve-process.js, ../src/commands/init.js, ../src/commands/create.js	13.09	~2.9% of CI-median 458s (mixed-machine estimate)
24	@objectstack/cli	packages/cli/test/scaffold-workspace-consistency.test.ts	packages/create-objectstack/src/templates/blank/pnpm-workspace.yaml, scripts/check-cross-package-test-inputs.mjs	PRIVATE internals: ../src/commands/init.js	2.68	~0.6% of CI-median 458s (mixed-machine estimate)
25	@objectstack/cli	packages/cli/test/serve-capability-identity.test.ts	packages/connectors/connector-mcp/src/connector-mcp-plugin.ts	PRIVATE internals: ../src/commands/serve.js	12.18	~2.7% of CI-median 458s (mixed-machine estimate)
26	@objectstack/cli	packages/cli/test/unbuilt-workspace-lead.test.ts	scripts/cross-package-test-inputs.mjs, scripts/cli-unbuilt-workspace-lead.mjs, scripts/cli-build-prerequisite.mjs, scripts/cli-unbuilt-workspace-lead.d.mts	PRIVATE internals: ../src/utils/invocation.js + repo path: scripts/cli-unbuilt-workspace-lead.mjs	0.09	~0.0% of CI-median 458s (mixed-machine estimate)
27	@objectstack/cli	packages/cli/test/vitest-tiers-partition.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./helpers/serve-process.js, ../vitest-tiers.js, ../vitest-tiers.fixtures.js + repo path: scripts/js-comment-mask.mjs	7.99	~1.7% of CI-median 458s (mixed-machine estimate)
28	@objectstack/client	packages/client/src/auth-route-ledger-coverage.test.ts	packages/plugins/plugin-auth/src/auth-route-ledger.conformance.test.ts, packages/plugins/plugin-auth/src/auth-route-ledger.ts	PRIVATE internals: ./index + repo path: packages/plugins/plugin-auth/src/auth-route-ledger	0.57	~1.8% of CI-median 32s (mixed-machine estimate)
29	@objectstack/client	packages/client/src/client-url-conformance.test.ts	packages/runtime/src/route-ledger.ts, packages/rest/src/rest-route-ledger.ts, packages/services/service-storage/src/storage-route-ledger.ts, packages/services/service-i18n/src/i18n-route-ledger.ts	PRIVATE internals: ./index + repo path: packages/runtime/src/route-ledger, packages/rest/src/rest-route-ledger, packages/services/service-storage/src/storage-route-ledger, packages/services/service-i18n/src/i18n-route-ledger, packages/plugins/plugin-auth/src/auth-route-ledger	3.74	~11.6% of CI-median 32s (mixed-machine estimate)
30	@objectstack/client	packages/client/src/envelope-caller-census.test.ts	scripts/****, scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs, scripts/†	repo path: scripts/js-comment-mask.mjs	14.16	~43.8% of CI-median 32s (mixed-machine estimate)
31	@objectstack/client	packages/client/src/rest-route-ledger-coverage.test.ts	packages/rest/src/rest-route-ledger.conformance.test.ts	PRIVATE internals: ./index + repo path: packages/rest/src/rest-route-ledger	0.67	~2.1% of CI-median 32s (mixed-machine estimate)
32	@objectstack/client	packages/client/src/route-ledger-coverage.test.ts	packages/runtime/src/route-ledger.conformance.test.ts	PRIVATE internals: ./index + repo path: packages/runtime/src/route-ledger	0.58	~1.8% of CI-median 32s (mixed-machine estimate)
33	@objectstack/client	packages/client/src/route-ledger-response-schema.test.ts	scripts/check-route-envelope.mjs, scripts/**	repo path: packages/runtime/src/route-ledger, packages/rest/src/rest-route-ledger, packages/services/service-storage/src/storage-route-ledger, packages/services/service-i18n/src/i18n-route-ledger, packages/plugins/plugin-auth/src/auth-route-ledger + @objectstack/spec/api	0.44	~1.4% of CI-median 32s (mixed-machine estimate)
34	@objectstack/client	packages/client/src/service-route-ledger-coverage.test.ts	packages/services/service-storage/src/storage-route-ledger.conformance.test.ts, packages/services/service-i18n/src/i18n-route-ledger.conformance.test.ts, packages/services/service-datasource/src/datasource-route-ledger.conformance.test.ts, packages/services/service-datasource/src/datasource-route-ledger.ts	PRIVATE internals: ./index + repo path: packages/services/service-storage/src/storage-route-ledger, packages/services/service-i18n/src/i18n-route-ledger, packages/services/service-datasource/src/datasource-route-ledger	0.68	~2.1% of CI-median 32s (mixed-machine estimate)
35	@objectstack/cloud-connection	packages/cloud-connection/src/canonical-expression-envelopes.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ./cloud-connection-ui.js, ./marketplace-ui.js + repo path: scripts/js-comment-mask.mjs	8.17	~24.3% of CI-median 34s (mixed-machine estimate)
36	@objectstack/cloud-connection	packages/cloud-connection/src/cloud-connection-route-ledger.conformance.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./cloud-connection-route-ledger.js + repo path: scripts/js-comment-mask.mjs	0.61	~1.8% of CI-median 34s (mixed-machine estimate)
37	@objectstack/core	packages/core/src/security/authz-store-unavailable.test.ts	packages//*.ts**	PRIVATE internals: ./resolve-authz-context.js, ./authz-store-unavailable.js	4.08	2.09% of 60s (local full suite, in-suite 1.25s)
38	@objectstack/core	packages/core/src/security/operation-private-keys.pin.test.ts	packages//*.ts**	public entry: @objectstack/core	0.80	1.87% of 60s (local full suite, in-suite 1.12s)
39	@objectstack/dogfood	packages/qa/dogfood/test/authz-conformance.test.ts	packages/rest/src/, packages/runtime/src/, packages/plugins/plugin-hono-server/src/, packages/services/service-realtime/src/, packages/client/src/, content/docs/protocol/kernel/realtime-protocol.mdx, packages/mcp/src/	PRIVATE internals: ./authz-conformance.matrix.js, ./authz-ledger-population.baseline.js	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
40	@objectstack/dogfood	packages/qa/dogfood/test/expression-conformance.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./expression-conformance.ledger.js	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
41	@objectstack/dogfood	packages/qa/dogfood/test/flow-trigger-conformance.test.ts	packages/spec/src/automation/**	PRIVATE internals: ./flow-trigger-conformance.ledger.js	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
42	@objectstack/dogfood	packages/qa/dogfood/test/packaged-activation-ledger-reach.dogfood.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	public entry: @objectstack/example-showcase, @objectstack/verify, @objectstack/connector-rest, @objectstack/connector-openapi, @objectstack/connector-mcp	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
43	@objectstack/dogfood	packages/qa/dogfood/test/route-ledger-live-mount-parity.dogfood.test.ts	packages/services/service-storage/src/storage-route-ledger.ts, packages/services/service-i18n/src/i18n-route-ledger.ts, packages/services/service-settings/src/settings-route-ledger.ts	repo path: packages/runtime/src/route-ledger.js, packages/rest/src/rest-route-ledger.js, packages/services/service-storage/src/storage-route-ledger.js, packages/services/service-i18n/src/i18n-route-ledger.js, packages/services/service-settings/src/settings-route-ledger.js + @objectstack/example-showcase, @objectstack/verify, @objectstack/service-storage	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
44	@objectstack/dogfood	packages/qa/dogfood/test/showcase-crud-persona-matrix.dogfood.test.ts	examples/app-showcase/**	PRIVATE internals: ./showcase-security.js	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
45	@objectstack/dogfood	packages/qa/dogfood/test/showcase-declarative-endpoints.dogfood.test.ts	packages/metadata/src/, packages/cli/src/commands/, packages/runtime/src/**	PRIVATE internals: ./build-shaped-artifact.js	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
46	@objectstack/dogfood	packages/qa/dogfood/test/showcase-declarative-mcp.dogfood.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	public entry: @objectstack/example-showcase, @objectstack/verify, @objectstack/connector-mcp, @objectstack/connector-openapi, @objectstack/connector-rest	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
47	@objectstack/dogfood	packages/qa/dogfood/test/validation-conformance.test.ts	packages/spec/src/data/**	PRIVATE internals: ./validation-conformance.ledger.js	NOT MEASURED	NOT MEASURED (dogfood: excluded from Test Core by --exclude, runs in its own file-sharded job; boots real apps)
48	@objectstack/downstream-contract	packages/qa/downstream-contract/test/source-resolution.pin.test.ts	packages/spec/package.json, packages/spec/src/†**	public entry: @objectstack/spec, @objectstack/spec/integration, @objectstack/spec/conversions	3.69	~37.3% of CI-median 10s (mixed-machine estimate)
49	@objectstack/driver-sql	packages/drivers/driver-sql/src/live-dialect-matrix.isolation.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ../src/index.js, ./live-dialect-matrix.testkit.js + repo path: scripts/js-comment-mask.mjs	8.78	~4.3% of CI-median 204s (mixed-machine estimate)
50	@objectstack/example-showcase	examples/app-showcase/test/inert-wirings.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ../objectstack.config.js, ../src/automation/jobs/index.js, ../src/security/bind-position-sets.js, ../src/security/demo-personas.js + repo path: scripts/js-comment-mask.mjs	11.57	~50.6% of CI-median 23s (mixed-machine estimate)
51	@objectstack/formula	packages/formula/src/rls-predicate.test.ts	packages/spec/src/security/rls.zod.ts	PRIVATE internals: ./rls-predicate, ./cel-to-filter, ./matches-filter	4.74	~38.3% of CI-median 12s (mixed-machine estimate)
52	@objectstack/formula	packages/formula/src/skill-catalog-sync.test.ts	skills/objectstack-formula/**	PRIVATE internals: ./validate	2.47	~19.9% of CI-median 12s (mixed-machine estimate)
53	@objectstack/lint	packages/lint/src/authoring-rule-wiring.test.ts	packages/metadata-protocol/src/, packages/cli/src/commands/†	PRIVATE internals: ./reference-integrity-suite.js, ./authoring-rules.js, ./runtime-gate.js	3.33	~3.8% of CI-median 87s (mixed-machine estimate)
54	@objectstack/lint	packages/lint/src/lint-startup-registry-verdict.corpus.test.ts	packages/services/service-automation/src/**	PRIVATE internals: ./lint-startup-registry-verdict.js	1.37	~1.6% of CI-median 87s (mixed-machine estimate)
55	@objectstack/lint	packages/lint/src/validate-jsx-pages.production-witness.test.ts	sdui.manifest.json, examples/app-showcase/src/ui/pages/capability-map.page.ts, examples/app-showcase/src/ui/pages/command-center-jsx.page.ts, examples/app-showcase/src/ui/pages/start-here.page.ts	PRIVATE internals: ./validate-jsx-pages.js, ./authoring-rules.js + repo path: examples/app-showcase/src/ui/pages/capability-map.page.js, examples/app-showcase/src/ui/pages/command-center-jsx.page.js, examples/app-showcase/src/ui/pages/start-here.page.js	1.68	~1.9% of CI-median 87s (mixed-machine estimate)
56	@objectstack/lint	packages/lint/src/validate-page-visualization-bindings.test.ts	examples/app-showcase/src/data/objects/task.object.ts, examples/app-showcase/src/ui/views/task.view.ts, examples/app-showcase/src/ui/pages/task-visualizations.pages.ts	PRIVATE internals: ./validate-page-visualization-bindings.js, ./reference-integrity-suite.js	1.27	~1.5% of CI-median 87s (mixed-machine estimate)
57	@objectstack/lint	packages/lint/src/validate-rule-compilability.test.ts	packages/objectql/src/validation/**	PRIVATE internals: ./validate-rule-compilability.js, ./authoring-rules.js	9.21	~10.6% of CI-median 87s (mixed-machine estimate)
58	@objectstack/lint	packages/lint/src/validate-translatable-sections.test.ts	examples/app-showcase/src/data/objects/contact.object.ts, examples/app-showcase/src/ui/views/contact.view.ts, examples/app-showcase/src/system/translations/index.ts	PRIVATE internals: ./validate-translatable-sections.js, ./validate-translation-references.js, ./showcase-shape.fixtures.js + repo path: examples/app-showcase/src/data/objects/contact.object.js, examples/app-showcase/src/ui/views/contact.view.js, examples/app-showcase/src/system/translations/index.js	0.96	~1.1% of CI-median 87s (mixed-machine estimate)
59	@objectstack/lint	packages/lint/src/validate-translation-references.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./validate-translation-references.js, ./showcase-shape.fixtures.js + repo path: examples/app-showcase/src/data/objects/contact.object.js, examples/app-showcase/src/ui/views/contact.view.js	4.92	~5.7% of CI-median 87s (mixed-machine estimate)
60	@objectstack/mcp	packages/mcp/src/canonical-expression-envelopes.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ./connect-ui.js + repo path: scripts/js-comment-mask.mjs	8.68	~25.2% of CI-median 34s (mixed-machine estimate)
61	@objectstack/metadata	packages/metadata/src/metadata-route-ledger.conformance.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ./metadata-route-ledger.js + repo path: scripts/js-comment-mask.mjs	0.31	~0.7% of CI-median 43s (mixed-machine estimate)
62	@objectstack/metadata-protocol	packages/metadata-protocol/src/sys-metadata-repository.draft-drain.test.ts	scripts/check-durability-degradation-log-level.mjs	PRIVATE internals: ./sys-metadata-repository.js	6.66	~2.4% of CI-median 279s (mixed-machine estimate)
63	@objectstack/objectql	packages/objectql/src/action-owner-key-single-source.test.ts	scripts/cross-package-test-inputs.mjs, packages//*.ts**	PRIVATE internals: ./action-governance.js	0.44	0.08% of 531s (local full suite, in-suite 0.45s)
64	@objectstack/platform-objects	packages/platform-objects/src/managed-api-method-affordance-sweep.test.ts	packages//*.object.ts**	public entry: @objectstack/lint	10.46	~55.7% of CI-median 19s (mixed-machine estimate)
65	@objectstack/platform-objects	packages/platform-objects/src/pages/canonical-expression-envelopes.test.ts	scripts/js-comment-mask.d.mts, packages/cloud-connection/src/cloud-connection-ui.ts, packages/lint/src/page-envelope-audit.test.ts, scripts/js-comment-mask.mjs	PRIVATE internals: ./index.js + repo path: scripts/js-comment-mask.mjs	9.20	~49.0% of CI-median 19s (mixed-machine estimate)
66	@objectstack/plugin-approvals	packages/plugins/plugin-approvals/src/manager-org-screen-parity.contract.test.ts	packages/plugins/plugin-sharing/src/team-graph.ts	PRIVATE internals: ./approval-service.js + repo path: packages/plugins/plugin-sharing/src/team-graph.js	8.59	~27.6% of CI-median 31s (mixed-machine estimate)
67	@objectstack/plugin-approvals	packages/plugins/plugin-approvals/src/status-mirror-cascade.integration.test.ts	scripts/check-cross-package-test-inputs.mjs, packages/runtime/src/expected-read-refusal-noise.ts	PRIVATE internals: ./approval-service.js, ./sys-approval-request.object.js, ./sys-approval-action.object.js, ./sys-approval-approver.object.js + repo path: packages/runtime/src/expected-read-refusal-noise.js	12.57	~40.4% of CI-median 31s (mixed-machine estimate)
68	@objectstack/plugin-auth	packages/plugins/plugin-auth/src/managed-extension-fields.test.ts	packages/core/src/security/, packages//*.object.ts†	PRIVATE internals: ./managed-extension-fields.js, ./sys-user-writable-fields.js	1.26	~0.6% of CI-median 222s (mixed-machine estimate)
69	@objectstack/plugin-auth	packages/plugins/plugin-auth/src/platform-owner-email-reader-census.pin.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	public entry: @objectstack/types	0.07	~0.0% of CI-median 222s (mixed-machine estimate)
70	@objectstack/plugin-auth	packages/plugins/plugin-auth/src/rate-limit-storage-isolation.test.ts	scripts/check-published-files.mjs, packages/types/src/node-isolation.test.ts, scripts/js-comment-mask.d.mts, scripts/check-cross-package-test-inputs.mjs, scripts/js-comment-mask.mjs, packages/runtime/src/, packages/services/service-sms/src/	repo path: scripts/js-comment-mask.mjs + @objectstack/plugin-auth/rate-limit-storage	0.76	~0.3% of CI-median 222s (mixed-machine estimate)
71	@objectstack/plugin-email	packages/plugins/plugin-email/src/transports/smtp-port-contract.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ./smtp-port-contract.js, ./smtp.js + repo path: scripts/js-comment-mask.mjs	7.02	~17.9% of CI-median 39s (mixed-machine estimate)
72	@objectstack/plugin-security	packages/plugins/plugin-security/src/audience-anchor-set-claims.pin.test.ts	packages/spec/src/identity/, packages/spec/src/security/	PRIVATE internals: ./objects/default-permission-sets.js	2.05	~1.8% of CI-median 115s (mixed-machine estimate)
73	@objectstack/rest	packages/rest/src/meta-state-route-doc-spelling.test.ts	content/docs/protocol/objectql/state-machine.mdx, content/, skills/objectstack-automation/SKILL.md, **skills/, docs/†	PRIVATE internals: ./rest-route-ledger.js	0.42	0.11% of 338s (local full suite, in-suite 0.37s)
74	@objectstack/runtime	packages/runtime/src/action-owner-key-single-source.test.ts	scripts/cross-package-test-inputs.mjs, packages//*.ts**	PRIVATE internals: ./action-execution.js	9.81	0.43% of 570s (local full suite, in-suite 2.44s)
75	@objectstack/runtime	packages/runtime/src/error-envelope.conformance.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ./http-dispatcher.js, ./error-envelope.js, ./dispatcher-error-vocabulary.js + repo path: scripts/js-comment-mask.mjs	11.11	0.55% of 570s (local full suite, in-suite 3.13s)
76	@objectstack/service-settings	packages/services/service-settings/src/value-domains.shared-predicate.pin.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ./zz-alpha2-array.js + repo path: scripts/js-comment-mask.mjs	0.38	~0.9% of CI-median 42s (mixed-machine estimate)
77	@objectstack/spec	packages/spec/scripts/category-title.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./lib/category-title	0.07	0.01% of 737s (local full suite, in-suite 0.07s)
78	@objectstack/spec	packages/spec/scripts/check-generated-ledger.test.ts	scripts/**	none (fs/child_process reads only)	3.34	0.44% of 737s (local full suite, in-suite 3.27s)
79	@objectstack/spec	packages/spec/scripts/dist-freshness-adoption.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	none (fs/child_process reads only)	16.05	2.06% of 737s (local full suite, in-suite 15.17s)
80	@objectstack/spec	packages/spec/scripts/dist-freshness.test.ts	scripts/**	PRIVATE internals: ./lib/dist-freshness + repo path: scripts/check-regen-pending.mjs	6.01	0.98% of 737s (local full suite, in-suite 7.23s)
81	@objectstack/spec	packages/spec/scripts/escape-mdx.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./lib/escape-mdx	0.11	0.02% of 737s (local full suite, in-suite 0.11s)
82	@objectstack/spec	packages/spec/scripts/export-list.test.ts	skills/**	PRIVATE internals: ./lib/export-list	0.05	0.01% of 737s (local full suite, in-suite 0.06s)
83	@objectstack/spec	packages/spec/scripts/file-description.test.ts	content/docs/references/**, **content/**, scripts/	PRIVATE internals: ./lib/file-description, ./realtime-shared.zod, ../shared/lazy-schema, ../shared/protection.zod, ../shared/retired-key, ../shared/strict-object, ../kernel/metadata-protection.zod, ../shared/expression.zod, ./artifact.zod	0.60	0.09% of 737s (local full suite, in-suite 0.68s)
84	@objectstack/spec	packages/spec/scripts/gen-sdui-manifest-cleanup.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	none (fs/child_process reads only)	3.82	0.48% of 737s (local full suite, in-suite 3.53s)
85	@objectstack/spec	packages/spec/scripts/gen-sdui-manifest-collision.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	none (fs/child_process reads only)	2.98	0.40% of 737s (local full suite, in-suite 2.98s)
86	@objectstack/spec	packages/spec/scripts/gen-sdui-manifest-write-target.test.ts	scripts/**	none (fs/child_process reads only)	0.45	0.06% of 737s (local full suite, in-suite 0.47s)
87	@objectstack/spec	packages/spec/scripts/liveness/evidence.test.ts	packages/runtime/src/, packages//*.ts, packages/objectql/src/validation/, packages/metadata-protocol/src/, packages/plugins/plugin-audit/src/**	PRIVATE internals: ./evidence.mts	0.76	0.11% of 737s (local full suite, in-suite 0.83s)
88	@objectstack/spec	packages/spec/scripts/liveness/proof-registry.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./proof-registry.mts	0.11	0.02% of 737s (local full suite, in-suite 0.14s)
89	@objectstack/spec	packages/spec/scripts/publish-smoke-port-collision.test.ts	scripts/***, packages/cli/src/commands/serve.ts, packages//.ts	none (fs/child_process reads only)	6.68	0.90% of 737s (local full suite, in-suite 6.66s)
90	@objectstack/spec	packages/spec/scripts/query-pointer-row.test.ts	content/docs/references/**, content/**	PRIVATE internals: ./lib/file-description	0.05	0.01% of 737s (local full suite, in-suite 0.08s)
91	@objectstack/spec	packages/spec/scripts/references-banner.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./lib/generated-output	0.05	0.01% of 737s (local full suite, in-suite 0.06s)
92	@objectstack/spec	packages/spec/scripts/root-index.test.ts	content/docs/references/**, content/**	PRIVATE internals: ./lib/sharded-artifacts, ./lib/root-index	0.09	0.01% of 737s (local full suite, in-suite 0.10s)
93	@objectstack/spec	packages/spec/scripts/schema-tree-freshness.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	repo path: scripts/check-regen-pending.mjs	0.11	0.01% of 737s (local full suite, in-suite 0.11s)
94	@objectstack/spec	packages/spec/scripts/strictness-ledger-doc.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./lib/strictness-ledger, ./lib/strictness-ledger-doc	7.90	0.74% of 737s (local full suite, in-suite 5.46s)
95	@objectstack/spec	packages/spec/scripts/strictness-ledger.test.ts	docs/audits/2026-07-unknown-key-strictness-ledger.md	PRIVATE internals: ./lib/strictness-ledger	1.49	0.23% of 737s (local full suite, in-suite 1.70s)
96	@objectstack/spec	packages/spec/src/api/error-catalog-docs.test.ts	content/docs/api/error-catalog.mdx, content/**	PRIVATE internals: ./errors.zod	0.10	0.01% of 737s (local full suite, in-suite 0.08s)
97	@objectstack/spec	packages/spec/src/data/api-methods-batch-conformance.test.ts	packages//*.object.ts†**	PRIVATE internals: ./api-derivation	0.34	0.04% of 737s (local full suite, in-suite 0.31s)
98	@objectstack/spec	packages/spec/src/identity/position-delegatable-enforcer.pin.test.ts	packages/lint/src/**	none (fs/child_process reads only)	0.11	0.01% of 737s (local full suite, in-suite 0.08s)
99	@objectstack/spec	packages/spec/src/shared/retired-key-migrate-sentence.test.ts	packages/lint/src/, packages//*.ts, .claude/skills/spec-property-retirement/SKILL.md	none (fs/child_process reads only)	1.43	0.22% of 737s (local full suite, in-suite 1.61s)
100	@objectstack/spec	packages/spec/src/system/compliance-families-retirement.test.ts	scripts/****, packages//.mts†, packages/**/.cts†, packages//*.js†, packages//.mjs†, packages/**/.cjs†, packages//*.json†, packages//.md†, packages/**/.mdx†, packages//*.yaml†, packages//.yml†, examples/**/.json†, examples//*.md†, examples//.mdx†, examples/**/.yaml†, examples/**/*.yml†	PRIVATE internals: ../../scripts/lib/export-origins-testkit, ../migrations/registry, ./index, ./incident-response.zod	5.97	0.40% of 737s (local full suite, in-suite 2.94s)
101	@objectstack/spec	packages/spec/src/system/constants/platform-object-names.test.ts	packages//*.object.ts†**	PRIVATE internals: ./platform-object-names	0.11	0.01% of 737s (local full suite, in-suite 0.10s)
102	@objectstack/spec	packages/spec/src/ui/action-requires-confirmation-docblock.pin.test.ts	packages/runtime/src/, packages//*.ts	none (fs/child_process reads only)	0.05	0.01% of 737s (local full suite, in-suite 0.04s)
103	@objectstack/trigger-api	packages/triggers/trigger-api/src/trigger-api-route-ledger.conformance.test.ts	scripts/js-comment-mask.d.mts, scripts/js-comment-mask.mjs	PRIVATE internals: ./plugin.js, ./trigger-api-route-ledger.js + repo path: scripts/js-comment-mask.mjs	0.35	~20.4% of CI-median 2s (mixed-machine estimate)
104	@objectstack/trigger-record-change	packages/triggers/trigger-record-change/src/record-change-integration.test.ts	scripts/check-cross-package-test-inputs.mjs, scripts/slot-lookup-baseline.json, packages/core/src/kernel.ts, packages/runtime/src/expected-read-refusal-noise.ts	PRIVATE internals: ./plugin.js + repo path: packages/runtime/src/expected-read-refusal-noise.js	14.66	~62.3% of CI-median 24s (mixed-machine estimate)
105	@objectstack/types	packages/types/src/driver-error-classification.callers.test.ts	packages//*.ts**	none (fs/child_process reads only)	1.87	32.09% of 18s (local full suite, in-suite 5.69s)
106	create-objectstack	packages/create-objectstack/src/banner-version.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./banner.js	0.87	~5.6% of CI-median 16s (mixed-machine estimate)
107	create-objectstack	packages/create-objectstack/src/scaffold-description.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	none (fs/child_process reads only)	3.86	~24.9% of CI-median 16s (mixed-machine estimate)
108	create-objectstack	packages/create-objectstack/src/scaffold-e2e-boot-probe.test.ts	scripts/gen-sdui-manifest.sh, scripts/publish-smoke.sh, packages/cli/src/commands/serve.ts, .github/workflows/scaffold-e2e.yml	none (fs/child_process reads only)	13.51	~87.2% of CI-median 16s (mixed-machine estimate)
109	create-objectstack	packages/create-objectstack/src/scaffold-next-steps-pm.test.ts	packages/cli/src/commands/init.ts	none (fs/child_process reads only)	3.40	~21.9% of CI-median 16s (mixed-machine estimate)
110	create-objectstack	packages/create-objectstack/src/scaffold-skills-single-copy.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	PRIVATE internals: ./skills-install.js	1.13	~7.3% of CI-median 16s (mixed-machine estimate)
111	create-objectstack	packages/create-objectstack/src/starter-comments-self-contained.test.ts	(reads via a loop variable / dir listing; radius held by a sibling row)	none (fs/child_process reads only)	0.06	~0.4% of CI-median 16s (mixed-machine estimate)
112	create-objectstack	packages/create-objectstack/src/template-consistency.test.ts	scripts/sync-template-versions.mjs, content/†**	PRIVATE internals: ./pkg-utils.js, ./template-copy.js, ./template-registry.js, ./skills-install.js	0.73	~4.7% of CI-median 16s (mixed-machine estimate)
113	create-objectstack	packages/create-objectstack/src/template-version-stamps.test.ts	scripts/invoked-as.mjs†	PRIVATE internals: ./scripts/sync-template-versions.mjs, ./invoked-as.mjs	0.25	~1.6% of CI-median 16s (mixed-machine estimate)

Generated by Claude Code

---

Generated by Claude Code

[claude]

os-dev-report

{
 "issue": 16466,
 "status": "needs_decision",
 "branch": "claude/issue-16466-repo-conformance-task",
 "pr": null,
 "premise_still_valid": true,
 "summary": "Ruling step 1 executed in full and step 2 decided by the table (posted as the comment above this one): 27 packages / 113 escaping test files (the registry holds 168 glob rows; the card's '117 rows' is the gate's self-test case count), 104 files measured with a per-module collect+run reporter, the six heavy packages' full suites measured locally for in-suite shares (spec 7.3%, core 4.0%, types 32.1% of a 13s suite, runtime 1.0%, objectql 0.1%, rest 0.1% — the value of a move is the cacheability of the other 93-99.9%, not the moved seconds). The count is under the ~60 bar (heavy six: 33 files) but the separability condition FAILS: 20 of 33 import PRIVATE internals of their home package, and the holders of every wide glob on spec (compliance-families-retirement.test.ts, which needs the unpublished scripts/lib/export-origins-testkit), runtime (standaloneActionObjectName / standaloneActionOwnerKey, not exported from @objectstack/runtime) and rest (REST_ROUTE_LEDGER, not exported from @objectstack/rest) cannot import through a public entry — verified against the built dist/*.d.ts. Per the ruling's 'otherwise' branch no edit was made, no PR opened (the branch exists, empty, as the claim's landing mark); the premise itself is confirmed and quoted: a docs-only diff moves spec#test, rest#test and create-objectstack#test hashes and --union-into pulls exactly those three into an otherwise empty merge-group set (~656s CI-median wall), a leaf .ts diff pulls spec+core+types+runtime+objectql (~911s). Separable as they stand: core 2 + types 1 + objectql 1 files (every private binding provably public) and spec 10 files that hold no wide glob the compliance pin does not also hold. Assignee was empty on arrival (PM-side half-state); the claim comment names this session and branch. Coordinator heads-up (X's turbo.json overlap, V's ci.yml landing) needs no action while this card ships no diff; V's landing was read and is what makes the docs-only reading above exact.",
 "tests": "Tree a9d25d842f (BASE); origin/main 4500bc5d79 by the end — only ci.yml differs among measured files. Build: `turbo run build --concurrency=2` under os-verify-lock, `Tasks: 73 successful, 73 total · Cached: 56 cached` (5m35s shared-box). Enumeration: `node scripts/check-cross-package-test-inputs.mjs --list-escapes` exit 0 → 27 packages, 113 files. Import-style + glob attribution by scripts in the scratchpad importing the gate's own findEscapingPackages()/coversDirectory and glob-match.mjs (literals map lit→test; heldBy witnesses added). Durations: vitest 4.1.10, per-module diagnostic() (collect+setup+prepare+env+run) via a custom reporter, `pnpm exec vitest run --maxWorkers=2 --reporter=default --reporter=REPORTER_PATH FILES` per package under the lock (slot issue-16466), every leg's exit echoed: spec 26 files exit 0 (390 tests), core 2/0, types 1/0, runtime 2/0, objectql 1/0, rest 1/0, group A 19 packages all exit 0 (172s), cli 27 files exit 0 (`Test Files 27 passed (27) · Tests 442 passed`, 127s); dogfood 9 files NOT MEASURED (excluded from Test Core, own job). Full suites: spec `--shard=1/3|2/3|3/3` all exit 0 (`Tests 4544/4118/4440 passed`, 70/113/344s), runtime exit 0 (`Test Files 238 passed · Tests 3361 passed`, 307.69s), objectql exit 0 (`284 passed · 4800 passed`, 290.88s), rest exit 0 (`187 passed · 3179 passed`, 186.74s), core pass 1 exit 0 (1232 passed) / pass 2 exit 1 (`1 failed | 1231 passed` — `Test timed out in 5000ms` in service-resolution-discriminator.contract.test.ts:227, filed #16497), types exit 0 (578 passed). Baseline gates on the untouched tree, exits captured before any pipe: `pnpm check:cross-package-test-inputs` exit 0 → `OK: 27 package(s) read outside themselves, all declared, and turbo.json hashes every declared glob.` (+ `All 117 self-test cases passed`); `node scripts/partition-test-shards.mjs --self-test` exit 0 → `self-test OK (71 measured packages → 72 shard items, 6 shards, max/mean 1.00x at most 1.3x, floor 404s ...)`; `pnpm check:nul-bytes` exit 0 → `OK (scanned 8083 text file(s) ...)`. Before-readings: three `turbo run test --dry-run=json` (clean / docs-only scratch edit / leaf scratch edit) compared by task hash — docs-only moves spec#test, rest#test, create-objectstack#test; leaf moves connector-slack, example-showcase, dogfood + spec, core, types, runtime, objectql; both scratch edits restored with `git checkout HEAD --` under an EXIT/INT/TERM trap with absolute paths, mutation proven on disk by grep -c = 1 and a moved blob hash, restore proven by hash-object == HEAD blob (yes/yes), `git diff HEAD --stat` and `git status --porcelain` both empty. `--union-into` from an empty package list: docs → `+ @objectstack/spec, + @objectstack/rest, + create-objectstack`; leaf → spec, core, types, objectql, runtime. No dispatch-gates derivation (no diff), no pnpm lint (no diff), no changeset (no PR). Not measured: cache HIT/MISS as such (no turbo test cache exists locally; hash movement is the invariant), any 'after' reading.",
 "mcp_calls": "0 — card and comments via the public-repo payload channel; dedup via REST list (8 pages, 674 open issues, page 8 returned 3 so the listing is complete) + local grep; finding filed via REST POST",
 "open_questions": [
  {
   "question": "The ruled shape (one repo-conformance vitest project owning all 33 heavy-package scanning files) is entangled: the wide-glob holders on spec, runtime and rest import private internals that no public entry exposes. Which shape, if any, should the follow-up dispatch execute?",
   "options": [
    "A. Full move as ruled (33 files → packages/qa/repo-conformance). Requires the new project to import spec's private scripts/lib/* testkits, runtime's action-execution.ts and rest's rest-route-ledger.ts by relative cross-package paths, and its registry rows to name those private files. Axes: real need yes (docs-only merge groups run spec+rest today, quoted); long-term poor — a qa package reading three packages' private modules inverts the dependency direction and is the 'no workarounds' shape; AI-error-proofing worse — private cross-package relative imports are the spelling the alias gate cannot see and the registry gate only flags; startup cost large (33 files, 6 turbo tasks, registry rewrite, 16 spec script tests re-pathed).",
    "B. Separable partial move inside the ruling: 4 files (core/src/security/authz-store-unavailable.test.ts, core/src/security/operation-private-keys.pin.test.ts, types/src/driver-error-classification.callers.test.ts, objectql/src/action-owner-key-single-source.test.ts) → packages/qa/repo-conformance, imports rewritten to the same symbols on @objectstack/core / @objectstack/objectql (verified present in dist/index.d.ts), their rows re-homed (core, types, objectql leave the registry; the new project declares packages/**/*.ts + scripts/cross-package-test-inputs.mjs), the three packages' #test tasks drop packages/**/*.ts; new package weighed by the fallback (countTestFiles × 1.522s) until the next timings refresh; runtime's twin keeps its literal path to the moved objectql twin updated (one string). Effect: a leaf .ts diff stops re-running core+types+objectql (243s of 911s CI-median, ~27%); docs-only merge groups unchanged (spec+rest still ~640s). Axes: real need partial — misses the maintainer's measured pain (docs-only 15-20-min shards); long-term clean (public entries only); AI-proofing neutral-to-good (the gate keeps red-by-default over the new home); startup cost small. Do NOT also move spec's 10 separable files: they drop no glob while the compliance pin stays, so it would be churn with zero cache effect.",
    "C. B plus rest: export REST_ROUTE_LEDGER from @objectstack/rest's public entry (api-surface + changeset: a published-surface widening) and move meta-state-route-doc-spelling.test.ts too, so rest#test drops content/**, docs/**, skills/**. Effect on docs-only merge groups: 656s → ~419s (spec + create-objectstack remain). Runtime's two non-exported bindings could be exported the same way for another 264s on the leaf case. Axes: real need yes for the rest half; long-term acceptable only if the maintainer wants those names public — otherwise it publishes internals for CI's convenience; AI-proofing neutral; startup cost small-plus-contract-tier.",
    "D. A different mechanism, not the ruled shape (ci.yml and the gate change): keep every file where it is and split each heavy package's `test` into two turbo tasks over two vitest projects in the same package — `test` (package-local inputs; an explicit include list that excludes the escaping files) and `test:repo` (the escaping files; the wide $TURBO_ROOT$ inputs) — with check-cross-package-test-inputs Layer B verifying the radius on `PKG#test:repo` and that every escaping file is in that project's include list (a mechanical, self-testable rule), Test Core running `test test:repo` per package, and the partitioner weighing test:repo by the fallback. No import changes, no private exposure, no file moves. Effect: docs-only merge groups re-run only the repo projects (spec ~54s + rest ~0.4s + create-objectstack ~24s instead of ~656s); leaf .ts diffs likewise. Axes: real need yes — the only option that reaches the measured docs-only pain; long-term sustainable (contract stays 'a file that reads outside its package is in the repo project or the gate reds'); AI-proofing good (same red-by-default posture, one more mechanical rule); startup cost moderate (gate rule + self-test, ci.yml, six vitest configs, turbo.json) and it needs a maintainer ruling because the card fixed ci.yml as untouched."
   ],
   "recommendation": "D as the direction, because on the first axis it is the only option that reaches what the maintainer actually measured (docs-only merge groups running 15-20-minute shards: today that is spec+rest+create-objectstack by the quoted union), on the second it keeps every private import where it belongs instead of moving it across a package boundary, on the third it adds one mechanical gate rule rather than a tolerated cross-package private path, and on the fourth it is a bounded change to two files the repo already owns as CI mechanism. If the maintainer prefers to stay inside the ruled shape and land something now, B is the honest increment (four files, public entries only, ~27% of the leaf-case cost, 0% of the docs-only case) — and C's rest half is worth adding only if REST_ROUTE_LEDGER is a name the maintainer is willing to publish. A is not recommended on any axis."
  }
 ],
 "out_of_scope_findings": [
  "filed as #16497: core's service-resolution-discriminator pin awaits a dynamic import of the whole package index under vitest's default 5000ms timeout and timed out once in two full-suite runs on a shared box (5036ms); sibling class of #16434"
 ]
}

Generated by Claude Code

---

Generated by Claude Code

[claude]

Ruling at the 04:3xZ delivery (report 5565121423, table 5565114830): option D as the direction. The card's ruled shape (one repo-conformance project owning the 33 files) is withdrawn on the measured separability failure: 20 of 33 files import private internals of their home package, and the wide-glob holders on spec, runtime and rest cannot import through any public entry. The card's "ci.yml untouched" line was the seat's own serial constraint against #16453 and #16455, not a maintainer ruling; it is lifted for the Test Core run step only, under the proofs below. Authority: the maintainer's direction 「同意你的建议，你负责执行派发所有可行的优化」 (live PM chat, 2026-09-07).

Shape:

1. Every file stays where it is. Each heavy package (spec, core, types, runtime, objectql, rest) splits its test into two vitest projects and two turbo tasks: test (package-local inputs; an explicit include list that excludes the escaping files) and test:repo (the escaping files; the wide $TURBO_ROOT$ inputs move there). check-cross-package-test-inputs gains the mechanical Layer-B rule: the radius is verified on PKG#test:repo, and every escaping file is in that project's include list; self-test cases in both directions (an escaping file left in test reds; a test:repo row without its wide input reds).
2. Precondition, measured before the ci.yml edit: whether Test Core in CI restores a turbo cache for test tasks at all (a cache step, TURBO_TOKEN / TURBO_TEAM, or nothing). If it does, the shard runs test test:repo for the selected packages and the cache carries the unmoved PKG#test. If it does not, hash movement alone gains nothing and the selection becomes task-level: the union step of scripts/ci/select-shard-packages.sh emits the moved tasks (spec#test:repo, not spec) and the shard runs exactly those tasks. State which case holds, with the reading, in the PR body.
3. The partitioner weighs test:repo by the fallback until the next timings refresh (ci: a scheduled workflow regenerates scripts/test-shard-timings.json from the newest fully green main run and opens the PR (maintainer-directed, measurement loop) #16464). The shard attestation and the completeness guard still count every file: prove by the before and after collected-file counts per package (no file lost, no file counted twice).
4. Serial: .github/workflows/ci.yml and turbo.json overlap ci: e2e and live tiers move to a nightly run on main; PR and queue runs keep unit, integration and conformance (maintainer-directed, part B tiering) #16455 (PR ci(test): e2e and live filename tiers move to a nightly run on main #16481, one push pending) on the same step and the same task overrides; scripts/ci/select-shard-packages.sh is ci: the merge queue runs the affected package set, not the full list (maintainer-directed, part A of the test-cost programme) #16453's (merged 04:04:38Z, on main). Branch from origin/main; before the final push prove git merge-tree --write-tree clean against PR ci(test): e2e and live filename tiers move to a nightly run on main #16481's then-current head in the form the brief prescribes; if it is not clean, stop at the draft PR and report blocked (the seat lands this card after ci: e2e and live tiers move to a nightly run on main; PR and queue runs keep unit, integration and conformance (maintainer-directed, part B tiering) #16455). ci: every Test Core run publishes the slowest test files and packages beside their pinned weights (maintainer-directed, part B measurement) #16454 stays behind both.
5. Acceptance, replacing the card's: the card's ablation stands in the new shape (a scratch docs-only commit on the PR branch: turbo run test test:repo --dry-run=json shows only the test:repo tasks of spec and rest plus create-objectstack#test moving; a scratch leaf .ts commit: the heavy six's test hashes do not move); both scratch commits restored and never pushed; the PR's own CI shows each shard's N/M items line and its wall for a package-touching diff. Clause-② yes stays; contract tier; changeset per the gate.
6. Not in scope: exporting REST_ROUTE_LEDGER or runtime's two bindings (option C), a published-surface widening nobody asked for; option B's four-file move, churn with no docs-only effect. Both recorded here, not filed.

The flight resumes on this ruling in the same session, same branch, same claim.

---

Generated by Claude Code