[objectstack-fleet]

Requested by Tesla · project thread

Part of #25677
Part of #25711

Clause-②: no
Metadata-protocol: none

Follow-up to #25745 (4a), per ruling 30 = B: the two Test UI / Test Core reds that #25745 caused once it merged are fixed here, on the same card. Test-only; nothing is released, so this carries skip-changeset.

What changes

1. Test UI shard 3/16: ui/packages/cli/src/__tests__/app-generator.test.ts. The release-lockstep test read ui/.changeset/config.json, which #25745 deleted (Changesets reads one config per repository). It now reads the config through changesetConfigPath() from ui/scripts/workspace-root.mjs, the helper every other reader of the Changesets config already uses: the tree's own config when standalone, the monorepo root's when nested under ui/.

2. Test Core 5/8: packages/lint/src/validate-jsx-pages.production-witness.test.ts. The witness asserted that the SDUI manifest contains the html component. objectui#13449 retired html, and the manifest #25745 regenerated from the in-tree ui/ has 106 components, without it. The witness now pins the html tier by two of its intrinsic tags (a, p). The shipped pages never used the raw html component, and the clean-run witness in the same file still checks them.

Verification (this branch, on main 2a61f08)

• app-generator.test.ts: 48/48 pass. Without fix 1: 1 failed / 47 passed (ENOENT on ui/.changeset/config.json).
• Test UI shard 3/16, built the CI way (the 7-package turbo build, then pnpm --dir ui run test --shard=3/16): 361 files passed, 1 skipped; 4,724 tests passed, 102 skipped.
• validate-jsx-pages.production-witness.test.ts: 5/5 pass. Also green: @objectstack/sdui-parser 225/225, the cli sdui test files 79/79, app-showcase metric-filter 7/7.

🤖 Generated with Claude Code

https://claude.ai/code/session_0149BHSXfCdQywGaLdpRAtWe

[github-actions]

📓 Docs Drift Check

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

What this run could not see

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

Coarse fallback — 0 page(s) merely mention a changed package (the pre-#9192 predicate, kept for the deliberately-wide backstop): node scripts/docs-audit/affected-docs.mjs --json 90a34e2039b7e374dadb92881905409ad491eff4 → packageMentionDocs.

[objectstack-fleet]

Gate verdict: head 6015709368 (2026-10-26T00:20Z). The head is main 90a34e2 (PR B #25827) merged in; the PR's own diff is 3 files, all test or test-gate data.

Since the PR opened, bd959d5757 fixed the Lint UI red. The changesetConfigPath() import moved the read registered in SUBJECT_IS_THE_CWD (ui/scripts/check-test-path-roots.mjs) from line 1326 to 1329, so the register now names the new line. It is the same entry with the same reason, and check:test-path-roots exits 0.

Local, this head:

• dispatch-gates --commands derives 59 families for the 3 paths. --ran reconciles: 59 derived, 59 run, 0 NOT-MEASURED, 0 UNRUN. 58 exit 0.
• The one red is check:dts-closure: @object-ui/components is missing dist/index.d.ts, because its declaration build hit 3 type errors in this environment (combobox.tsx, resolve-icon.ts, input-otp.tsx). This diff touches none of those files, and Type Check UI and Build & E2E UI are green on main in CI.
• ui/ scripts/__tests__, the whole suite: the only reds are the 12 nested ui/.github tests below. Four files that timed out under load (check-hand-rolled-comment-mask, check-vi-mock-inherit, dollar-dialect-alias-census, js-comment-mask-returns-interpolation-9424) pass when re-run alone, 178/178 together with check-test-path-roots.test.ts.
• app-generator.test.ts 48/48. Test UI shard 3/16 built the CI way: 4,724 tests passed. validate-jsx-pages.production-witness.test.ts 5/5.

Known reds on main that are not this PR's:

• Test UI 8/16 + 9/16 on main are the nested ui/.github class (ci-cd-pipeline-doc.test.ts, workflow-cache-save-bound.test.ts), owned by tooling(ui-merge): governance in-repo — liveness evidence and spec tests stop citing objectui by commit, one sdui-parser, AGENTS.md / pm scripts drop objectui#, ui/AGENTS.md nests, 18 governance workflows merged or retired (parent #25677, 4b/6) #25712 (next card).
• Lint · repo & tooling gates + Lint · PM dispatch-gates cleared with PR B ci(ui-merge): retire the two-repo pin gates — Spec Shape Gate, Console Pin Gate, prerelease-pin-watch #25827, which is now on main and merged into this head.

[github-actions]

⛔ merge queue 构建失败 — 先分诊，再决定要不要重排

队列构建 40251505048 红了。队列跑的是全量套件（PR 侧 CI 只跑 affected 子集），
所以失败的测试可能在本 PR 没碰过的包里 —— 那不是重排能修的。每次盲目重排都会让排在后面的所有 PR 重建一轮。

分类：failure —— 按下面的日志分诊。

失败的 job（日志抽取，best effort）：

• Test UI (8/16) — 失败步骤: Run the ui tests (shard 8/16)

  × gives every workflow in .github/workflows/ a row in the inventory table — in both directions 15ms
  × never names a workflow file that does not exist 7ms
  × documents every job these four workflows define 2ms
  × names every first-party command the four jobs actually run 2ms
  × names no command outside the declared non-`run:` set 3ms
  × reads one section per workflow — a shrunken enumeration is a broken reader 3ms
  × finds commands on both sides — a silent parser would make every pin below vacuous 4ms
  × names every first-party command every workflow job actually runs 21ms
  × declares every command a section names that its workflow does not run 3ms
  × finds no pre-install gate reaching a package 9ms
  FAIL   unit  scripts/__tests__/check-pre-install-import-graph.test.ts > the real tree — every step objectui#6148 measured is still derived > finds no pre-install gate reaching a package
    ↳ 失败原因: AssertionError: a script this repository runs before `pnpm install` needs `node_modules` to load: expected [ Array(1) ] to deeply equal []
  FAIL   unit  scripts/__tests__/ci-cd-pipeline-doc.test.ts > ci-cd-pipeline.md — workflow inventory > gives every workflow in .github/workflows/ a row in the inventory table — in both directions
    ↳ 失败原因: AssertionError: the "Workflow Inventory" table has rows for workflows that are NOT in .github/workflows/:
  

• Test UI (9/16) — 失败步骤: Run the ui tests (shard 9/16)

  × bounds the three jobs objectui#7270 found running under the 360-minute default 4ms
  × bounds the four jobs objectui#7956 derived a ceiling for 1ms
  FAIL   unit  scripts/__tests__/workflow-cache-save-bound.test.ts > every workflow cache save is bounded and non-fatal (objectui#7048) > bounds the three jobs objectui#7270 found running under the 360-
    ↳ 失败原因: AssertionError: changeset-release.yml must still define a `release:` job: expected undefined to be defined
  FAIL   unit  scripts/__tests__/workflow-cache-save-bound.test.ts > every workflow cache save is bounded and non-fatal (objectui#7048) > bounds the four jobs objectui#7956 derived a ceiling for
    ↳ 失败原因: AssertionError: changeset-release.yml must still define a `lane:` job: expected undefined to be defined
  

↳ 失败原因 是判读的关键：超时（Test timed out in … / Hook timed out in …）多半是负载/时序，不是本 PR 的回归；
断言（AssertionError: …）才指向真实的行为改变。两者的 FAIL 行长得一模一样，只有这一行能区分。

⚠️ 断言这一侧有一类例外，判据是断言在测什么，不是它是不是 AssertionError。 断言的对象是产品行为（一个值、一个形状、一次拒收）⇒ 照上面读：真实的行为改变，去查，⛔ 不要重排掉；
断言的对象是这次实验自身的有效性前提（跑完的耗时、负载下的先后、任何只在时间预算内才成立的条件）⇒ 它跟超时是同一类，同样对负载敏感，重排一次是合法的判别手段。
识别是机械的：断言的消息或它比较的值本身点名了一段时长、一个时间戳、一个耗时计数。实测过的一对 —— AssertionError: SecurityPlugin.init() ran: expected false to be true 测的是产品行为（真回归）；
AssertionError: this run took over a second, so second-precision stamps could have differed too: expected 1006 to be less than 1000 测的是实验前提：它守护的那条不变式当时是绿的，同一个 head 原样重排一次即成功。
穿着 AssertionError 外衣的时间测量，仍然是时间测量。（⛔ 这只改「怎么读一次红」，不改「哪些测试可以重排」——后者由别处管。）

跨 PR 相同签名（24h，按失败测试文件聚合）：

• scripts/__tests__/check-pre-install-import-graph.test.ts — 24h 窗口内只有本 PR 撞到过，暂不汇总（再有一个不同 PR 撞到就会自动开汇总 issue）。
• scripts/__tests__/ci-cd-pipeline-doc.test.ts — 24h 窗口内只有本 PR 撞到过，暂不汇总（再有一个不同 PR 撞到就会自动开汇总 issue）。
• scripts/__tests__/workflow-cache-save-bound.test.ts — 24h 窗口内只有本 PR 撞到过，暂不汇总（再有一个不同 PR 撞到就会自动开汇总 issue）。

历史信号：

• 本 PR 过去 24h 无队列失败记录（首次）。
• 过去 24h 队列共有 3 个失败构建（不含本次）。

分诊清单：

1. 失败测试在本 PR 改动的包里 → 真回归，修 PR。
2. 失败测试与本 PR 无关 → 看上面的「跨 PR 相同签名」；已有汇总 issue ⇒ flaky/环境问题实锤，去那张 issue 上谈，修好前重排只会再烧一轮全队列。
3. 两者都不是 → 可能与同组 PR 语义冲突；等前面的 PR 落地或失败出队后再重排一次即可，不要连续重排。

---

Generated by Claude Code · merge-queue-triage workflow (#4859)