test(objectql,verify,runtime): declare OS_REGISTRY_LOG=warn in each suite's own vitest harness by zhuangjianguo · Pull Request #14016 · objectstack-ai/objectstack

zhuangjianguo · 2026-08-31T23:58:33Z

Part of #13517 — item 1 of the three-part remainder that card carries. It stays open: the packages/rest stack-dump residual and the deferred log-level ratchet are still on it, and the structured-logger half is #13986.

b79ddf17df quieted packages/qa/dogfood with one declarative OS_REGISTRY_LOG=warn in that suite's own vitest harness. This applies the same seam to the other three app-booting suites — packages/objectql, packages/verify, packages/runtime — each in its own config, each with its own measurement in the docblock.

Measured per suite, not inferred from dogfood

One full pnpm --filter PKG test per suite per side, all six runs through the shared verify lock. Baseline at origin/main b1b7d6088a; after-side at this branch's head 0fc8e11c18.

suite	stdout lines	`[Registry]` lines	console-carried lines	test outcome
`objectql` before	16,194	4,744	5,189	250 files / 4,322 tests passed
`objectql` after	11,502	65	510	250 files / 4,322 tests passed
`verify` before	5,670	2,323	2,551	9 files / 48 tests passed
`verify` after	3,347	0	228	9 files / 48 tests passed
`runtime` before	6,730	1,155	1,964	202 files / 3,011 tests passed
`runtime` after	5,584	9	818	202 files / 3,011 tests passed

"console-carried" excludes the structured logger's TIMESTAMP INFO ... lines, which never go through console and are a different question (#13986). Test outcomes are identical on both sides of every suite, and the non-[Registry] line counts are identical to the line — 445 / 228 / 809 before and after — so this key moved exactly the family it names and nothing else.

The zero in the verify row is a real reading, not an unfired grep: the same pattern on the same suite's baseline log counts 2,323, and grep -c "Test Files" on the after-side log answers 1.

Where the route does NOT do what dogfood did

⚠️ packages/runtime is not a second dogfood, and its docblock says so. [Registry] is 59% of that suite's console-carried output, not dogfood's 94.9%. The ~800 lines left are a mixed tail no log level reaches: [sql-driver] while creating/syncing table ... column reports, [HonoServerPlugin] Server stopped, Paged read of ... is NOT deterministic, [action-audit] ..., [Protocol] DB hydration skipped .... The declaration still earns its line there (1,155 lines removed), and it is also not the whole story there.

objectql (91.5%) and verify (91.3%) do behave like dogfood.

One repair the declaration made necessary, named here because it is a test edit

packages/objectql/src/registry-log-level.test.ts is the pin for how OS_REGISTRY_LOG resolves — so this package's harness key lands on that file's own subject. The file already deleted the variable in afterEach; the symmetric beforeEach was missing, so the FIRST case ran at warn under a name that says info, and stayed green, because both of its expectations are negative ("not called").

Measured, not reasoned. With a premise assertion added and the beforeEach withheld:

AssertionError: expected 'warn' to be 'info'
 Test Files  1 failed (1)
      Tests  1 failed | 6 passed (7)

With the beforeEach in place: 2 passed (2) / 19 passed (19) for that file plus registry-collision-order.test.ts. Both "default level" cases now assert the level they actually run at, so no future harness key can silently re-point them. No test was skipped, disabled or re-baselined; two assertions were added and one hook body was widened. No rebuild leg applies to that measurement — the file imports ./registry by relative path, so vitest transforms the source and no dist is in the path.

The near-miss, measured rather than trusted

registry-collision-order.test.ts asserts on [Registry] Collision lines. Those go through a bare console.warn the level never gates, and the file sets logLevel = 'silent' per registry anyway. Both sides pass identically, and the after-side logs still carry the collision lines — 16 in objectql, 2 in runtime. A real shadowing still speaks.

What this deliberately is not

The engine's shipped default is untouched — still 'info' at packages/objectql/src/registry.ts:1265 — and no library code learns what a test runner is. The request lives in each harness, declaratively, where the test author can see it. check-console-intercept-disarm.mjs is green: the disableConsoleIntercept line is untouched in all three configs and the new key sits beside it.

Verification

All gates run at 0fc8e11c18, exit codes captured before any pipe.

pnpm lint (repo-wide eslint . --no-inline-config) — exit 0, no narrowing claimed
pnpm check:nul-bytes — exit 0, 7667 text file(s) ... no raw ASCII control bytes
26 of the 27 families named by node scripts/pm/dispatch-gates.mjs --repo objectstack-ai/objectstack, plus check-console-intercept-disarm — all exit 0
pnpm --filter typecheck on all three packages — exit 0
NOT MEASURED here, owned by CI, all three exit 3 = PREREQUISITE NOT MET rather than a finding: check-test-completeness (wants a saved turbo run test log), check:dual-build-cjs-loads and check:type-check-debt (want the whole workspace built)
packages/objectql's tsc program excludes **/*.test.ts, so its green typecheck says nothing about the edited test file — verified with --listFiles (0 test files in the program, registry.ts present as the control). The edited file was type-checked directly instead, in a throwaway project over the package's own options: 0 errors, file present in the program.

Generated by Claude Code

…uite's own vitest harness Measured per suite on origin/main b1b7d60, one full run each: objectql 16,194 stdout lines · 4,744 [Registry] · 10,984 structured-logger verify 5,670 stdout lines · 2,323 [Registry] · 3,113 structured-logger runtime 6,730 stdout lines · 1,155 [Registry] · 4,760 structured-logger Against the console-carried population each suite emits (the structured logger writes to process.stdout directly and is a separate question), [Registry] is 91.5% in objectql, 91.3% in verify and 59.0% in runtime. OS_REGISTRY_LOG is @objectstack/objectql's own published seam for that verbosity (SchemaRegistryOptions.logLevel / REGISTRY_LOG_LEVELS). The engine's shipped default stays 'info', and no library code learns what a test runner is: the request lives in each harness, declaratively. Claude-Session: https://claude.ai/code/session_01F3jdziLbAPGeceVNmSox5L Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

… about The package harness now declares OS_REGISTRY_LOG=warn for every worker, and this file's subject IS that variable. Its `afterEach` already deleted the var; the symmetric `beforeEach` was missing, so the FIRST case ran at 'warn' under a name that says 'info' — and stayed green, because both of its expectations are negative ("not called"). Measured rather than reasoned: with the premise assertion added and the `beforeEach` withheld, the case fails with `expected 'warn' to be 'info'`. Both "default level" cases now assert the level they run at, so a future harness key cannot silently re-point them again. Claude-Session: https://claude.ai/code/session_01F3jdziLbAPGeceVNmSox5L Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

github-actions · 2026-08-31T23:59:01Z

📓 Docs Drift Check

⚠️ 3 changed file(s) yielded no anchor (packages/objectql/vitest.config.ts, packages/runtime/vitest.config.ts, packages/verify/vitest.config.ts), so the pages documenting them are NOT COVERED by this run — this is not a clean bill of health for those files. Nothing else in this diff resolved to a documentable surface (no symbol, route or SDK anchor derived from 3 changed package(s)).

What this run could not see

3 changed file(s) yielded no anchor (packages/objectql/vitest.config.ts, packages/runtime/vitest.config.ts, packages/verify/vitest.config.ts) — pages documenting those are invisible to this run
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.

Coarse fallback — 33 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 62a137baecb75d03c028779ef5c2ed0eacad8399 → packageMentionDocs.

`packages/rest` is the untouched CONTROL in docs/audits/2026-09-test-log-volume-census.md: while PR #13985 and PR #14016 declared a registry log level in four harnesses, this suite got none, and the audit measures 528 residual `[Registry]` lines here — eight times what declared objectql still emits. Six test files in this package construct bare `SchemaRegistry` instances, so the chatter is per registered item per case. This declares the level in the harness only. objectql's shipped `'info'` default is unchanged for every production reader, and no library code is made aware of a test runner: `OS_REGISTRY_LOG` is the engine's own published seam (`SchemaRegistryOptions.logLevel` / `REGISTRY_LOG_LEVELS`). The ADR-0005 `[Registry] Collision` diagnostics ride a bare `console.warn` that the level never gates, so a real shadowing still speaks here. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012zGPuVVX3deAx9LdjK8jCk

`packages-write-envelope.test.ts` constructs bare `SchemaRegistry` instances per case and installs a package manifest into them, so this suite pays a `[Registry] Registered …` line per registered item at the engine's shipped `'info'` default. Harness-only, the same shape PR #14016 landed for three other suites: objectql's shipped default stays `'info'` for every production reader, and no library code learns it is under a test runner. `OS_REGISTRY_LOG` is the engine's own published seam. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012zGPuVVX3deAx9LdjK8jCk

…ting suites objectql's `SchemaRegistry` logs one `[Registry] Registered <kind>: <name>` line per registered item per construction, at its shipped `'info'` default. A suite that boots app stacks pays items x boots of that on a CI shard log nobody reads. PR #13985 and PR #14016 declared `OS_REGISTRY_LOG` in four harnesses and removed ~47,900 console-carried lines — and nothing held them there. Deleting one `env` line restored five figures of output while every assertion stayed green and every outcome was unchanged. This gate asserts the declaration. It reuses `check-console-intercept-disarm`'s exported `workspacePackageDirs` walk and its verdict format, so only the assertion and the population are new. THE POPULATION IS DERIVED, not a roster. A vitest-running package is selected when its OWN TEST SOURCES construct a `SchemaRegistry`, boot a stack through `@objectstack/verify`'s `bootStack`, or import an `@objectstack/example-*` app. On this tree that selects 8 of 72 with zero over- and zero under-selection. A roster of the four suites that already declared would have been a hand list this repo could show to be wrong on the day it shipped: docs/audits/2026-09-test-log-volume-census.md measures `packages/rest` — its own named control — at 528 residual `[Registry]` lines, eight times declared objectql's 66. The four packages the predicate newly selects adopt the declaration in the four commits preceding this one. Comment masking is what makes the predicate honest rather than merely plausible: `packages/cli` names `bootStack` and `@objectstack/verify` in eight places, all of them prose, and would otherwise be selected while booting no stack in any test. The registry-construction signal additionally masks string and template content, so a spelling inside a literal is never the setting. The level vocabulary is READ from objectql's own `REGISTRY_LOG_LEVELS` rather than copied, because `registry.ts` resolves an unrecognised value to the `'info'` DEFAULT silently — `OS_REGISTRY_LOG: 'quiet'` would read as a considered choice and quiet nothing. Out of scope and untouched: which level any suite picks, objectql's shipped `'info'` default, and any library awareness of a test runner. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012zGPuVVX3deAx9LdjK8jCk

…t, not ~45% (objectstack-ai#14989) * wip(qa): log-volume census instrument Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 * wip(qa): log-volume census checkpoint — 32/72 packages measured Extends Batch 1's ledger (23 packages, driven under the shared verify lock) with 9 more: driver-sqlite-wasm, driver-turso, formula, lint, mcp, metadata, metadata-core, metadata-fs, metadata-protocol. Also: end-to-end control for the OS_LOG_LEVEL premise re-verification (Zone 2 D) — resolveLogLevel() confirmed sensitive to its input, and a repo-wide grep confirms it is read nowhere outside packages/cli (one non-hit is a self-test fixture string in dispatch-gates.mjs). Doc's method/mechanism/reproduction/premise sections carried over from a prior attempt's draft, independently re-verified against source (packages/core/src/logger.ts, packages/verify/src/harness.ts) before being kept — every file:line citation checked out exactly. Provisional structured share (console+structured) at 32/72: ~84%, already far from the five-suite ~45%, in the direction the earlier reading's own explanation predicts. Not the final number — continuing. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 * wip(qa): log-volume census checkpoint — 45/72 packages measured +13 packages this batch: observability, platform-objects, embedder-openai, knowledge-memory, knowledge-ragflow, plugin-approvals, plugin-audit, plugin-auth, plugin-dev, plugin-email, plugin-hono-server, plugin-pinyin-search, plugin-reports. plugin-auth is the first non-trivial plugins/ package: 4,840 lines (2,149 structured / 2,678 console) — most plugins so far are near- silent (13 reporter lines only, no kernel boot in their own test script). Provisional structured share (console+structured) at 45/72: ~80%. 27 remain, all services/*, triggers/*, spec, sdui-parser, and the qa/http-conformance + qa/downstream-contract pair. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 * wip(qa): log-volume census checkpoint — 55/72 packages measured +10 packages this batch: plugin-security, plugin-sharing, plugin-webhooks, downstream-contract, http-conformance, sdui-parser, service-analytics, service-automation, service-cache, service-cluster. 17 remain: service-cluster-redis, service-datasource, service-i18n, service-job, service-knowledge, service-messaging, service-package, service-queue, service-realtime, service-settings, service-sms, service-storage, spec, trigger-api, trigger-record-change, trigger-schedule, types. Provisional structured share (console+structured) at 55/72: ~79%, holding steady across three checkpoints now (84% -> 80% -> 79%). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 * wip(qa): log-volume census checkpoint — 68/72 packages measured +13 packages this batch: service-cluster-redis, service-datasource, service-i18n, service-job, service-knowledge, service-messaging, service-package, service-queue, service-realtime, service-settings, service-sms, service-storage, spec. 4 remain: trigger-api, trigger-record-change, trigger-schedule, types. packages/spec measured read-only (its own `test` script run and stdout captured, nothing edited under it — domain:spec's package stays untouched); 481s of the 480s budget, exit 0. Provisional structured share (console+structured) at 68/72: ~78%. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 * docs(qa): finalize log-volume census — 72/72 packages, ratio does not hold All 72 workspace packages with a test script measured, all green. Final totals: 61,980 structured / 17,428 console / 993 reporter / 80,401 total lines — structured share of total 77.1% (comparable to the earlier five-suite framing, whose own `console` bucket absorbed what this instrument tracks separately as `reporter`). The five-suite ~45% figure does not hold, and not toward more noise: it moved to 77.1% structured, decomposed in "The answer" section as two effects — (1) two commits (objectstack-ai#13985, objectstack-ai#14016) that landed on main between the earlier reading and this one already took the original five suites from ~45% to 86.6% structured on their own, by cutting a `[Registry]` debug-console population; (2) extending to the other 67 packages pulls that back down to 77.1% (not further), since that population's own structured share (57.7%) is lower than the five heaviest suites' but is also a minority (30.3%) of total test-run volume, and most of it (35/72 packages under 30 lines total, 22 of those exactly 13 -- pure reporter boilerplate) never boots a kernel at all. No seam added -- per triage's ruling (comment 5489397004), this document is the measurement; which seam (if either) to build is triage's call. Merged origin/main (dispatch-gates warned STALE TREE, 19 commits behind) -- no relevant file (logger.ts, harness.ts, log-level.ts, vitest configs, package.json test scripts) changed across that range, so the merge does not invalidate any measurement already taken. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 * docs(qa): note 85 direct new ObjectKernel() call sites, not just harness.ts Relevant to the fork this card hands back to triage: a BootOptions field on packages/verify's harness would only quiet suites that boot through that harness, not the ~60 other files (85 call sites) that construct a kernel directly -- including packages/objectql's own kernel-factory.ts, which alone accounts for 18.3% of the structured lines measured. Does not choose a seam; changes what each seam would cost to actually reach the population it's meant to quiet. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 * docs(qa): fix three citation defects in the log-volume census Review found three citation defects in docs/audits/2026-09-test-log-volume-census.md — the measurement itself (72-row table, arithmetic) was independently reproduced and is untouched. 1. `new ObjectKernel(` repo-wide count: re-ran the exact grep command on this branch merged with origin/main and got 78 occurrences across 55 files, not 85/62. Corrected the number and the two downstream figures that depend on it (~53 remaining call sites, "all 78 call sites"), and stamped the ref it was verified against so a future drift is reproducible instead of silent. 2. `scripts/pm/dispatch-gates.mjs:14461` is blank on this tree; the real self-test-fixture line is 14478. Corrected and ref-stamped (this file is edited often). 3. `packages/core/src/logger.ts:350-395` — the quoted block actually starts at 349, not 350 (content verbatim-correct, only the start line was off by one; end line shifts by the same one to 394). Re-resolved the other eight citations the reviewer already checked (logger.ts:180/236, harness.ts:96-313/384, kernel.ts:87, log-level.ts:53, kernel-factory.ts:35, and the 72/72 disableConsoleIntercept:true claim) against this branch's merged tree — none of them moved. Part of objectstack-ai#13986 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 * docs(qa): bump verified-at ref stamps to the post-second-merge tree origin/main moved again (99b4deb) between the first fix commit and the gate sweep. Re-verified all three corrected citations plus the eight already-checked ones against that tree — nothing moved — and updated the inline ref stamps from 48d4422 to 0de50b8 (this branch's HEAD after merging origin/main a second time) so they point at the tree the numbers were actually last confirmed on. Part of objectstack-ai#13986 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0112hMx9hjJ9BgB28X97DS68 --------- Co-authored-by: Claude <noreply@anthropic.com>

…bjectstack-ai#15559) * test(rest): declare OS_REGISTRY_LOG in the vitest harness `packages/rest` is the untouched CONTROL in docs/audits/2026-09-test-log-volume-census.md: while PR objectstack-ai#13985 and PR objectstack-ai#14016 declared a registry log level in four harnesses, this suite got none, and the audit measures 528 residual `[Registry]` lines here — eight times what declared objectql still emits. Six test files in this package construct bare `SchemaRegistry` instances, so the chatter is per registered item per case. This declares the level in the harness only. objectql's shipped `'info'` default is unchanged for every production reader, and no library code is made aware of a test runner: `OS_REGISTRY_LOG` is the engine's own published seam (`SchemaRegistryOptions.logLevel` / `REGISTRY_LOG_LEVELS`). The ADR-0005 `[Registry] Collision` diagnostics ride a bare `console.warn` that the level never gates, so a real shadowing still speaks here. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012zGPuVVX3deAx9LdjK8jCk * test(client): declare OS_REGISTRY_LOG in the vitest harness `packages-write-envelope.test.ts` constructs bare `SchemaRegistry` instances per case and installs a package manifest into them, so this suite pays a `[Registry] Registered …` line per registered item at the engine's shipped `'info'` default. Harness-only, the same shape PR objectstack-ai#14016 landed for three other suites: objectql's shipped default stays `'info'` for every production reader, and no library code learns it is under a test runner. `OS_REGISTRY_LOG` is the engine's own published seam. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012zGPuVVX3deAx9LdjK8jCk * test(service-automation): declare OS_REGISTRY_LOG in the vitest harness `connector-reload-reingest.test.ts` and `flow-name-shadowing.test.ts` construct bare `SchemaRegistry` instances and register connectors and flows into them per case, so the registration chatter is per item per case at the engine's shipped `'info'` default. `flow-name-shadowing.test.ts` asserts on shadowing warnings, which ride a bare `console.warn` this level never gates — the suite is measured green with the declaration in place. Harness-only: objectql's shipped default is untouched. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012zGPuVVX3deAx9LdjK8jCk * test(app-showcase): declare OS_REGISTRY_LOG in the vitest harness This config's own `disableConsoleIntercept` docblock deferred the question in as many words: "most of it `[Registry]` registration chatter. Quieting THAT is a separate question about `@objectstack/objectql`'s own default log level, not about this setting." This answers it the way the four earlier harnesses did — the level is declared HERE, in the harness, through the engine's own `OS_REGISTRY_LOG` seam. `gap-fill.test.ts` constructs the registry it reaches. objectql's shipped `'info'` default is unchanged for every production reader. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012zGPuVVX3deAx9LdjK8jCk * feat(devx): ratchet the declared registry log level across engine-booting suites objectql's `SchemaRegistry` logs one `[Registry] Registered <kind>: <name>` line per registered item per construction, at its shipped `'info'` default. A suite that boots app stacks pays items x boots of that on a CI shard log nobody reads. PR objectstack-ai#13985 and PR objectstack-ai#14016 declared `OS_REGISTRY_LOG` in four harnesses and removed ~47,900 console-carried lines — and nothing held them there. Deleting one `env` line restored five figures of output while every assertion stayed green and every outcome was unchanged. This gate asserts the declaration. It reuses `check-console-intercept-disarm`'s exported `workspacePackageDirs` walk and its verdict format, so only the assertion and the population are new. THE POPULATION IS DERIVED, not a roster. A vitest-running package is selected when its OWN TEST SOURCES construct a `SchemaRegistry`, boot a stack through `@objectstack/verify`'s `bootStack`, or import an `@objectstack/example-*` app. On this tree that selects 8 of 72 with zero over- and zero under-selection. A roster of the four suites that already declared would have been a hand list this repo could show to be wrong on the day it shipped: docs/audits/2026-09-test-log-volume-census.md measures `packages/rest` — its own named control — at 528 residual `[Registry]` lines, eight times declared objectql's 66. The four packages the predicate newly selects adopt the declaration in the four commits preceding this one. Comment masking is what makes the predicate honest rather than merely plausible: `packages/cli` names `bootStack` and `@objectstack/verify` in eight places, all of them prose, and would otherwise be selected while booting no stack in any test. The registry-construction signal additionally masks string and template content, so a spelling inside a literal is never the setting. The level vocabulary is READ from objectql's own `REGISTRY_LOG_LEVELS` rather than copied, because `registry.ts` resolves an unrecognised value to the `'info'` DEFAULT silently — `OS_REGISTRY_LOG: 'quiet'` would read as a considered choice and quiet nothing. Out of scope and untouched: which level any suite picks, objectql's shipped `'info'` default, and any library awareness of a test runner. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012zGPuVVX3deAx9LdjK8jCk --------- Co-authored-by: Claude <noreply@anthropic.com>

…es' files outside src to the commits that decided them (objectstack-ai#20923) Part of objectstack-ai#20594 Clause-②: no ## What changed This is stage 16 of the `domain:cli` lane's dead-citation sweep: **the 17 comment sites left in eleven lane packages' files outside `src/**`** (claim `5917064266`; the list on stage 15's landing `5916232164`). The file surface is the one stage 14 set: `README.md`, `tsconfig*.json`, `vitest.config.*`, `tsup.config.*` and `test/**`. Every comment site on that surface whose tracker number answers 404 now cites the commit that decided what the line describes, in ruling C+D's form C (comment `5749154545` on objectstack-ai#19123). Stages 1 to 15 of this card are the precedents; the latest are PR objectstack-ai#20883 and PR objectstack-ai#20898. - **17 sites on 16 lines in 14 files**, covering **9 numbers**, now cite **9 distinct commits**. - The 14 files: 5 JSONC configs (`tsconfig.test.json` in plugin-hono-server, plugin-dev and verify; plugin-hono-server's `tsconfig.typecheck.json`; client's `tsconfig.json`), 6 `vitest.config.ts` (client, cloud-connection, mcp, rest, runtime, types) and 3 test files (`qa/vitest-filter-preflight` 2, `qa/downstream-contract` 1). - **Comments only.** 16 lines out and 16 in, and every file keeps its line count. A token guard (below) shows zero non-comment tokens changed. - **String literals, `describe` / `it` titles and messages are untouched** (they belong to objectstack-ai#20752, fold `5911936313`). - **No tracker number is added.** The live numbers that share a changed line stay as they were: objectstack-ai#4914, objectstack-ai#12542 and objectstack-ai#17978 all answer 200. - **Sites left without a deciding commit: none.** - Where an earlier stage anchored a number and the line here describes the same decision, the same anchor is reused: 6 of the 9 numbers. The other 3 anchors are new (marked below). ## Census **Instrument.** The card's gate does not read these files (its declared surface is `packages/**/src/**`), so the census is taken by hand with the gate's own grammar, as stages 14 and 15 took it: - **Files:** every tracked file of the eleven packages outside `src/**`, with `CHANGELOG.md` left out: 81 files. Each is classed ON stage 14's file list (48 files) or OFF it (33: `package.json`, `LICENSE`, ledgers, `objectstack.config.ts` and the like). Only ON-list comment sites are this stage's. - **Extraction:** `extractCitations` from `scripts/check-issue-citations.mjs`, with its comment-prose projection for code and JSONC files and the whole file for `*.md`. The whole-file extraction minus the comment projection gives the string sites. - **Numbers kept:** only those naming this repository (`namesThisRepository`). - **Probe:** each distinct number, `GET /repos/objectstack-ai/objectstack/issues/N`; 404 means dead. | | tree | probe window (UTC) | numbers | answer 200 | answer 404 | dead comment sites, on-list | comment sites | dead comment sites, off-list | dead string sites | |---|---|---|---|---|---|---|---|---|---| | before | base `cb4c31dd52` | 2026-09-30 18:18:37 to 18:19:15 | 94 | 80 | 14 | **17** (9 numbers, 14 files) | 212 | 2 | 11 | | after | head `1e3782200f` | 2026-09-30 18:23:26 to 18:24:03 | 88 | 80 | 8 | **0** | 195 | 2 | 11 | - The before count matches the list on stage 15's landing, package by package. - No number present in both probes changed its answer. The 6 numbers that left the census (objectstack-ai#8651, objectstack-ai#10485, objectstack-ai#12181, objectstack-ai#13176, objectstack-ai#15145, objectstack-ai#16917) were only on the rewritten lines. - Comment sites fell by exactly 17. The off-list and string counts did not move. **Per package, dead on-list comment sites, before to after:** | package | before | after | |---|---|---| | `packages/plugins/plugin-hono-server` | 4 | 0 | | `packages/plugins/plugin-dev` | 2 | 0 | | `packages/client` | 2 | 0 | | `packages/qa/vitest-filter-preflight` | 2 | 0 | | `packages/cloud-connection` | 1 | 0 | | `packages/mcp` | 1 | 0 | | `packages/qa/downstream-contract` | 1 | 0 | | `packages/rest` | 1 | 0 | | `packages/runtime` | 1 | 0 | | `packages/types` | 1 | 0 | | `packages/verify` | 1 | 0 | | **total** | **17** | **0** | ## Per-number anchors Each anchor was checked by blame on the site and in the anchor's own message or diff. "Reused" names earlier stages that gave the number the same anchor. | number | sites | anchor | what it decided | | |---|---|---|---|---| | `objectstack-ai#13176` | 4: plugin-hono-server and plugin-dev `tsconfig.test.json` (:3 and :61 / :56) | `a68c61267` | plugin-security's test layer enters tsc under a sibling `tsconfig.test.json` at zero residue, with no `test-typecheck-debt.json`; its diff writes the number into that config's header and into the TEST_DEBT graduation | reused (plugin-security) | | `objectstack-ai#11332` | 1: plugin-hono-server `tsconfig.typecheck.json:12` | `dce5cd4f0` | retires the manifest's `capabilities` / `configuration` / `extensions` containers as `retiredKey()` tombstones (ADR-0049); its changeset names the number | reused (spec) | | `objectstack-ai#10724` | 1: the same line | `be21955ba` | retires the nine dead `contributes` members as `retiredKey()` tombstones (ADR-0049); its subject names the number | reused (spec) | | `objectstack-ai#12181` | 2: client `tsconfig.json:8`, `vitest.config.ts:33` | `cf71d73f8` | `meta.deleteItem`'s reset carriers; this same commit wrote both blocks (the `paths` rules and the alias for the real-door test), and its changeset names the number | reused (client, cli) | | `objectstack-ai#17853` | 5: vitest-filter-preflight's two test headers, rest / runtime / types `vitest.config.ts` | `08f5f0e5a` | a vitest filter that selects no test file says so; its message names the card it lands. The sentence in the three configs is the one stage 14 and stage 15 rewrote in cli's and dogfood's | reused (qa, cli, dogfood) | | `objectstack-ai#16917` | 1: cloud-connection `vitest.config.ts:64` | `7ce3154e6` | the comment-only repair of this file, which wrote the "paraphrased rather than quoted" sentence; its message names the card it lands | **new** | | `objectstack-ai#8651` | 1: mcp `vitest.config.ts:18` | `8c65046e4` | pins mcp's three engine doubles to metadata-core's dispatch predicates, adds that devDependency and creates this config for the alias; its message names the card it lands | **new** | | `objectstack-ai#10485` | 1: downstream-contract `test/contract.test.ts:46` | `35ad101bc` | retires `ThemeSchema` and the `themes` carrier key (ADR-0049, kept beside it); its subject names the number, and it wrote this line | reused (spec, qa, cli) | | `objectstack-ai#15145` | 1: verify `tsconfig.test.json:1` | `45a72b0cc` | wires verify's test layer into `typecheck` through this file; its diff writes the number into this line and into the coverage-ledger graduation | **new** | **ADR and ruling records.** A grep of `docs/adr` and `scripts/adr-anchors` for the 9 numbers finds one hit: ADR-0088 names objectstack-ai#10724 in a correction note (history, not the ruling), so that number stays on the commit every earlier stage anchored it to. The sentence it sits in already names ADR-0049. The control number `7329` finds 1 file in the same tree. **Anchor checks:** - **Each sha is unambiguous:** `git rev-parse --disambiguate` gives count 1 for each of the 9. - **Each is a plain commit** with one parent. - **Each is on the base:** `merge-base --is-ancestor` against `cb4c31dd52` exits 0 for all 9. - **The history is complete:** the checkout is not shallow. Control leg: `01218124ae`, the parent of the oldest anchor `8c65046e4` (2026-08-16), exits 0. Negative control: the base as an ancestor of `8c65046e4` exits 1. ## Verification All at head `1e3782200f`. Heavy runs went through `scripts/pm/os-verify-lock.sh` with `OS_VERIFY_LOCK_SLOT=issue-20594-outside`. - **Build (whole workspace):** `pnpm exec turbo run build --filter='./packages/*' --filter='./packages/*/*' --concurrency=2` → 71 successful, 71 total (9 cached), lock verdict command-exit 0. - **Typecheck, ten packages** (cloud-connection declares no `typecheck` script): `pnpm --workspace-concurrency=2 --filter` plugin-hono-server, plugin-dev, client, vitest-filter-preflight, mcp, downstream-contract, rest, runtime, types, verify `run typecheck` → 10 of 10 `Done`, 0 `error TS`, lock verdict command-exit 0. That reads every touched JSONC config (plugin-hono-server's runs `tsconfig.typecheck.json` by name; the `tsconfig.test.json` files are read by `check:test-typecheck`, all at their recorded ledgers). `--listFilesOnly` holds all 3 touched test files in their packages' programs. - **Tests, by name** (one lock call, verdict command-exit 0). Each touched test file ran, and each touched `vitest.config.ts` was loaded by a run of its own package: - vitest-filter-preflight: `test/config-wiring-sweep.test.ts`, `test/filter-preflight.test.ts` → 2 files, 97 tests passed. - downstream-contract: `test/contract.test.ts` → 1 file, 13 tests passed. - client: `src/meta-delete-item-carriers.test.ts` (the suite the edited alias block serves) → 20 passed. - cloud-connection: `src/canonical-expression-envelopes.test.ts` (the suite the alias exists for) → 16 passed. - mcp: `src/mcp-stdio-tools.test.ts` (the engine doubles the edited block describes) → 12 passed. - rest, runtime, types (`--project local`): `src/rest-exec-ctx-memo.test.ts` 4 passed, `src/app-plugin.ordering.test.ts` 3 passed, `src/email-verified.test.ts` 2 passed. - **Token guard**, base `cb4c31dd52` against head, 14 files, 5,764 base tokens: **0 files differ**. TypeScript files are compared as leaf tokens (`getChildren`, JSDoc nodes skipped), JSONC files as the scanner's non-trivia token stream plus their parsed value (0 of 5 values differ). Controls, in memory only: a comment inserted into each file, 0 of 14 differ; a statement appended, 14 of 14; one character flipped inside each file's first string token, 14 of 14 (9 `StringLiteral`, 5 JSON strings). - **Control bytes:** a scan of the 14 files finds 0 (positive control, a scratch file holding U+0001: 1). `pnpm check:nul-bytes` exits 0. - **Nothing publishes.** `npm pack --dry-run` in each of the 9 public packages packs only `dist/**`, `README.md`, `CHANGELOG.md`, `LICENSE` and `package.json`: 0 of the 11 touched files in those packages is packed. After the build, the first 48 characters of each of the 16 added comment lines occur in 0 files of any package's `dist/`; control: a `src` docblock phrase an earlier stage wrote ("Maintainer-seat ruling, landed by commit cf71d73") occurs in `packages/client/dist/index.d.ts`. The other two packages are private. ## Gates All at head `1e3782200f`, exit codes captured before any pipe. - **Derivation:** `node scripts/pm/dispatch-gates.mjs --repo objectstack-ai/objectstack --commands`, no paths, derives **52 commands** (14 paths against merge base `cb4c31dd52`). Re-derived after fetching `origin/main` to `4d0b9cd542`: the same 52, and none of the 4 upstream commits touches a derivation input or a file here. - **All 52 exit 0**, each on its first run. - **Reconciliation:** `dispatch-gates --ran` over the recorded `COMMAND :: exit CODE` list → "52 derived, 52 run, 0 NOT-MEASURED, 0 UNRUN", exit 0. - **Roster gates the derivation marks as keeping a roster under `packages/`:** `pnpm check:authz-resolver`, `pnpm check:error-code-casing` and `pnpm check:filter-alias-parity` → exit 0 each. - **`pnpm lint`** (repo-wide `eslint . --no-inline-config`, the family the derivation does not name) → exit 0, 2026-09-30T18:26:35Z to 18:30:31Z. - `node scripts/check-issue-citations.mjs` (diff mode, in the 52) reads 0 files, because none of these paths is on its declared surface; the hand census above is the measurement for this surface. - A local `git merge-tree --write-tree` of this head against `origin/main` `4d0b9cd542` is clean. **The 52 derived commands:** stage 15's list with `pnpm --filter @objectstack/spec run check:skill-examples` and `node scripts/check-tenant-audit-census.mjs` (with its `--self-test`) in, and `check:empty-state`, `check:liveness`, `check:strictness-ledger` and `check:variant-docs` (spec) out: - `node scripts/check-ci-filter-parity.mjs` - `node scripts/check-closing-keyword-parity.mjs` and `--self-test` - `node scripts/check-comment-mask-adoption.mjs` and `--self-test` - `node scripts/check-comment-mask-corpus.mjs` - `node scripts/check-issue-citations.mjs` - `node scripts/check-keyed-text-bounds.mjs` and `--self-test` - `node scripts/check-platform-object-tenancy-census.mjs` and `--self-test` - `node scripts/check-plugin-teardown-shape.mjs` and `--self-test` - `node scripts/check-registry-log-declared.mjs` and `--self-test` - `node scripts/check-rest-log-spy-declared.mjs` and `--self-test` - `node scripts/check-system-context-census.mjs` and `--self-test` - `node scripts/check-tenant-audit-census.mjs` and `--self-test` - `node scripts/check-undeclared-dep-imports.mjs` and `--self-test` - `node scripts/docs-audit/check-affected-docs.mjs` - `node scripts/docs-audit/check-drift-comment.mjs` - `pnpm --filter @objectstack/spec run check:skill-examples` - `pnpm check:cross-package-test-inputs`, `check:dispatcher-error-vocabulary`, `check:doc-authoring`, `check:driver-memory-census`, `check:dts-closure`, `check:dual-build-cjs-loads`, `check:engine-double-contract`, `check:gitlink-declared`, `check:issue-citations`, `check:lean-entry-closure`, `check:logger-receiver-detach`, `check:nul-bytes`, `check:objectql-double-limit`, `check:org-identifier`, `check:page-declaration-shape`, `check:published-files`, `check:query-options-erasure`, `check:refd-timer-probe`, `check:slot-lookup`, `check:sourcemap-no-sources-content`, `check:test-source-alias`, `check:tier-file-adoption`, `check:type-check-coverage`, `check:type-check-debt`, `check:watch-hint-literal`, `check:where-matcher` ## Acceptance notes - **Changeset:** none. Nine of the eleven packages publish, but no touched file is in any package's `files` (measured above), and `qa/vitest-filter-preflight` and `qa/downstream-contract` are private. The PR carries `skip-changeset`. - **Left in these eleven packages, not this stage's:** 2 dead comment sites off stage 14's file list, in plugin-hono-server's `objectstack.config.ts` (:19 objectstack-ai#11332, :26 objectstack-ai#10724), and 11 dead string sites: the `test-typecheck-debt.json` notes in client (1), mcp (1), rest (7) and runtime (1), and vitest-filter-preflight's `package.json` description (1). All 13 are among the 55 off-list sites stage 14 listed. - **What remains on this card after this stage:** the 55 off-list sites stage 14 listed, and the five `cli` sites with no deciding commit (objectstack-ai#10149, objectstack-ai#11048, objectstack-ai#14874 x3). None of them is touched here. - **Grammar reach, measured on this surface:** the gate's grammar does not extract the second number of a slash-joined pair such as `objectstack-ai#10374/objectstack-ai#13522`. On these 81 files that shape holds 4 such numbers (objectstack-ai#3060, objectstack-ai#7778, objectstack-ai#13522, objectstack-ai#14016), and all 4 answer 200 (2026-09-30T19:04Z), so the census above misses no dead site. - **No open PR touches these files.** The claim's serial check read all 11 open PRs' file lists at `cb4c31dd52`; objectstack-ai#20602 and objectstack-ai#20583 on this seat stay in `packages/rest/src` and `packages/cli/src`. --- _Generated by [Claude Code](https://claude.ai/code/session_01VvcEokUG1tvVxkceYfR5XB)_ Co-authored-by: Claude <noreply@anthropic.com>

claude added 2 commits August 31, 2026 23:31

github-actions Bot added the size/m label Aug 31, 2026

claude Bot added the skip-changeset PR has no user-facing published change; bypasses the changeset gate label Aug 31, 2026

github-actions Bot added the tests label Aug 31, 2026

zhuangjianguo marked this pull request as ready for review September 1, 2026 00:17

zhuangjianguo enabled auto-merge September 1, 2026 00:17

zhuangjianguo added this pull request to the merge queue Sep 1, 2026

Merged via the queue into main with commit 5e2c04d Sep 1, 2026
37 checks passed

zhuangjianguo deleted the claude/issue-13517-remaining-suite-log-levels branch September 1, 2026 00:48

claude Bot mentioned this pull request Sep 4, 2026

[PM seat] domain:devx @ objectstack — 🟢 os-elon-musk · 由 os-elon-musk 账户的项目负责 #6023

Open

claude Bot mentioned this pull request Sep 4, 2026

Ratchet the declared registry log level across engine-booting suites #15559

Merged

objectstack-fleet Bot mentioned this pull request Sep 30, 2026

docs: re-anchor the last dead tracker citations in eleven lane packages' files outside src to the commits that decided them #20923

Merged

Navigation Menu

Repository navigation

test(objectql,verify,runtime): declare OS_REGISTRY_LOG=warn in each suite's own vitest harness - #14016

test(objectql,verify,runtime): declare OS_REGISTRY_LOG=warn in each suite's own vitest harness#14016
zhuangjianguo merged 2 commits into
mainfrom
claude/issue-13517-remaining-suite-log-levels

zhuangjianguo commented Aug 31, 2026

Uh oh!

github-actions Bot commented Aug 31, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

zhuangjianguo commented Aug 31, 2026

Measured per suite, not inferred from dogfood

Where the route does NOT do what dogfood did

One repair the declaration made necessary, named here because it is a test edit

The near-miss, measured rather than trusted

What this deliberately is not

Verification

Uh oh!

github-actions Bot commented Aug 31, 2026

📓 Docs Drift Check

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants