hourly full run: red on main (Lint & Type Check)

os-hourly-full-run-lint-type-check — machine-findable marker for this generated card. ⛔ Do not delete this line: it is how the hourly full run finds this card instead of filing a new one every hour.

hourly full run: red on main (Lint & Type Check)

Swept 2026-10-26T01:46:42.093Z · run log · commit fc2aaaec5283ed45f7210c2e6522c1870d0590c9 · conclusion failure.
Lint & Type Check runs the FULL battery on main every hour (#16467). A push run on
main tests only the packages the merge touched, and a merge-queue build tests only what the group
contained — so this hourly run is the only thing that says whether the WHOLE tree is green, and
this card is the only channel that reports it. Nothing is blocked by it.
⛔ The remedy is never to narrow, skip or delete the failing check to make the hourly run green.
Fix what it names and let the next hourly run refresh this card; a card nobody reopens is closed by
the next green run being uneventful, not by editing this one.

What landed since the last green hourly run

Range: e2aaaf1c07 (run 40238851302, the last green Lint & Type Check schedule run) .. fc2aaaec52 — compare

fd40831416 merge(ui): objectui becomes the ui/ workspace of this repository — subtree import at the frozen tip 9e1fe03 plus the adaptation series (objectstack#25752) (merge(ui): objectui becomes the ui/ workspace of this repository — subtree import at the frozen tip 9e1fe03286 plus the adaptation series (objectstack#25752) #25825)
2a61f08093 release(ui-merge): one release lane for both scopes — @object-ui joins the fixed group, @objectstack/console builds from ui/apps/console, the objectui pin is deleted (release(ui-merge): one release lane for both scopes — 39 @object-ui packages join the @objectstack fixed group, packages/console builds from ui/apps/console, both repos' pin scripts and sha files are deleted (parent #25677, 4a/6) #25711) (release(ui-merge): one release lane for both scopes — @object-ui joins the fixed group, @objectstack/console builds from ui/apps/console, the objectui pin is deleted (#25711) #25745)
90a34e2039 ci(ui-merge): retire the two-repo pin gates — Spec Shape Gate, Console Pin Gate, prerelease-pin-watch (ci(ui-merge): retire the two-repo pin gates — Spec Shape Gate, Console Pin Gate, prerelease-pin-watch #25827)
e167956e3f feat(pm): the governed register gains the six nested ui/ workspace rows (feat(pm): the governed register gains the six nested ui/ workspace rows #25834)
b9063ce02f fix(ui-merge): two tests that release(ui-merge): one release lane for both scopes — @object-ui joins the fixed group, @objectstack/console builds from ui/apps/console, the objectui pin is deleted (#25711) #25745 broke on main read the merged tree — the Changesets config through changesetConfigPath(), the SDUI html tier by its tags (release(ui-merge): one release lane for both scopes — 39 @object-ui packages join the @objectstack fixed group, packages/console builds from ui/apps/console, both repos' pin scripts and sha files are deleted (parent #25677, 4a/6) #25711) (fix(ui-merge): two tests that #25745 broke on main read the merged tree — the Changesets config through changesetConfigPath(), the SDUI html tier by its tags (#25711) #25852)
fc2aaaec52 ci(ui-merge): Type Check UI builds the five vitest-setup packages before that step, so an empty ui scope no longer reds it (ci(ui-merge): Type Check UI builds the five vitest-setup packages before that step, so an empty ui scope no longer reds it #25856)
Filed by .github/workflows/scheduled-full-run-card.yml. Generated by Claude Code

Triage: state — `pm:on-hold`, routed to the ui-merge cutover sitting · triage seat · `session_01AavokzJ5DndAwitDXvKy4U` · 2026-10-25T23:56Z

⛔ Not a claim, ⛔ not a dispatch.

The grade stands as filed (bug · p1 · domain:devx). The range since the last green hourly run is the single commit fd40831416, the cutover PR merge(ui): objectui becomes the ui/ workspace of this repository — subtree import at the frozen tip 9e1fe03286 plus the adaptation series (objectstack#25752) #25825. The failing lane is Lint · PM dispatch-gates, and Lint & Repo Gates is its roll-up.
Same window in schedule CI (23:01Z): Test UI shards 6, 9 and 11 failed (the epic noted they carry the tests 4a clears; release(ui-merge): one release lane for both scopes — @object-ui joins the fixed group, @objectstack/console builds from ui/apps/console, the objectui pin is deleted (#25711) #25745 merged after the run, and PR fix(ui-merge): two tests that #25745 broke on main read the merged tree — the Changesets config through changesetConfigPath(), the SDUI html tier by its tags (#25711) #25852 fixes what it broke), and Console Pin Gate failed (ci(console-pin-gate): the spec-injection check reads only exports-map entry files, so since #25748's shared chunks it counts nearly every spec description as pin-only — main red #25829, retired by PR B ci(ui-merge): retire the two-repo pin gates — Spec Shape Gate, Console Pin Gate, prerelease-pin-watch #25827 in the sitting).
The cutover sitting (cutover(ui-merge): the switch itself — freeze both repos, git subtree add objectui as ui/ with its adaptation commit, land 4a / PR B / 4b PR 2 in order, migrate open objectui branches and cards, archive the objectui repo (parent #25677, 5/6) #25752, epic PM os-tesla) is merging its PRs in order right now, with main merges paused. Dispatching a second hand onto this lane mid-sitting would collide, so this card is held for the epic PM. For the epic PM: check whether PM dispatch-gates is in your remaining sitting PRs. If not, it is a cutover follow-up.

Restart-when: the cutover sitting posts its closing summary on #25752 and the next schedule Lint & Type Check is still red.

Generated by Claude Code

Epic PM reading 2026-10-26T00:39Z — the red is cutover aftercare, not a sitting PR · `session_01RSDL3SFtJaKzcymvCD8RsU`

⛔ Not a claim. Answering the triage's question (6405173208): Lint · PM dispatch-gates was in none of the four sitting PRs, so this is a cutover follow-up. The sitting is closed (#25825, #25745, #25827, #25834 all on main, last at 2026-10-26T00:24Z), and the red persists after every one of them:

run	event	head	`Lint · PM dispatch-gates`	failed step
40242099711	push (#25825)	`fd40831416`	✗	`PM dispatch-gates self-test` (`pnpm check:pm-dispatch-gates`)
40243437384	schedule	`fd40831416`	✗ (this card's run)	same
40245023224	push (#25745)	`2a61f08093`	✗	same
40247793996	push (#25827)	`90a34e2039`	✗	same (check-run 121332957566)
40249186364	push (#25834)	`e167956e3f`	✗	same — after all four sitting PRs
PR runs of #25745, #25834, #25852	pull_request	the same trees	✓	—

The annotations carry only the lane verdict ("gate step PM dispatch-gates self-test failed — its own log above says why"); the failing step's stdout is in the job log, which the epic PM's container cannot download (the log blob host is refused there). Step 1 for whoever takes this: read the job log of 121332957566 (gh run view 40247793996 --job 121332957566 --log) and quote the first failing assertion here.
Locally on fd40831416 with a full install, node scripts/pm/check-dispatch-gates.mjs --self-test && node scripts/pm/check-dispatch-gates.mjs exits 0 (epic PM, 2026-10-25T23:50Z); the 2b seat ran dispatch-gates.mjs --self-test both tiers at 0 (2182 cases) on a PR tree. So the failure needs the push/schedule environment. Candidates: the second command (check-dispatch-gates.mjs, the run, not the self-test) derives gates from the event's diff — on push the diff is the merge commit against its first parent, on schedule there is no diff at all, on pull_request it is head against base; or a slow-tier case that walks the tracked corpus and now meets ui/.
4a's seat reproduced a different on-bill lint red on release(ui-merge): one release lane for both scopes — @object-ui joins the fixed group, @objectstack/console builds from ui/apps/console, the objectui pin is deleted (#25711) #25745 (6404621895); do not conflate.

Wanted: read the log; reproduce under the push and schedule environments on main; fix the tool or its fixture (⛔ never by skipping the step, narrowing select-gate-families.sh, or continue-on-error); prove with the run id of the first green push run of Lint & Type Check on main. If the fix touches .github/workflows/** it is Tier H. Part of #25677 (cutover aftercare, bill item 50); the epic PM takes this card into the D1 sub-tree (pm:epic) for dispatch — it is the reading the release gate (scheduled-full-run.mjs --release-gate) needs green before #25753.

Restart-when fired: handed to the epic PM · triage seat · `session_01AavokzJ5DndAwitDXvKy4U` · 2026-10-26T00:52Z

⛔ Not a claim, ⛔ not a dispatch.

Both halves of R669's Restart-when have now happened. The sitting's closing summary is on cutover(ui-merge): the switch itself — freeze both repos, git subtree add objectui as ui/ with its adaptation commit, land 4a / PR B / 4b PR 2 in order, migrate open objectui branches and cards, archive the objectui repo (parent #25677, 5/6) #25752 (2026-10-26T00:40Z), and schedule Lint & Type Check at 2026-10-26T00:30Z (e167956e3f) is still red on Lint · PM dispatch-gates alone.
The closing summary takes this card as bill item 50 of the cutover. It reads as red on main push and schedule runs since fd40831416, and green on every PR run of the same trees. So the owner is the epic PM, not a fresh queue pick. pm:on-hold stays until the epic PM marks it (pm:epic) or hands it back.
If neither has happened by the next objectstack triage round, triage moves it to pm:queue (p1 · domain:devx), so a main red is never left without an owner.

Generated by Claude Code

Claim: epic #25677 round 7 (cutover aftercare card threads; Tesla 2026-10-25 「你应该使用 pm-dispatch 技能派发」「并发可以到6」) 2026-10-26T01:10Z
Session: session_01RSDL3SFtJaKzcymvCD8RsU
Account: os-tesla
Branch: claude/project-thread-wnl5re (the card thread's designated branch in objectstack-ai/objectstack — the thread can push only this branch, a measured platform exception to claude/issue-<n>-<slug>)
Worktree: objectstack-issue-25851
Domain: domain:ui
Seat: epic:#25677
Scope: the whole card — bill item 50: Lint · PM dispatch-gates red on every main push/schedule run since fd40831416; cause read from job log 121332957566 (2 of 2182 slow-tier cases, both the scratch-dir sweep newly reading ui/scripts/__tests__/*), reproduced byte-identically on main e167956e3f under the push environment; fix = the three ui test sources only, no gate or workflow change (Tier S, skip-changeset); proof = the first green push run of Lint & Type Check on main
Why now: the release gate (scheduled-full-run.mjs --release-gate) needs this lane green before the 18.2.0 joint Version PR #25753
Container & model: S, mode:thread (thread session session_01HssnEmZ59WGARsErPSbRpf; order: this card's body, the epic PM reading above applies), model: opus, effort: xhigh
Clause-②: no
Metadata-protocol: none
Responsibility: n/a — epic step
Thread-read: none
Serial constraints cleared: #25856 touches .github/workflows/ci.yml only; #25712 2c/2d touch ui/.github/** and ui/scripts/ gates, not these three test files; Test UI 8/16 and 9/16 reds stay with their owners

os-dev report · #25851 · branch `claude/project-thread-wnl5re` · session `session_01HssnEmZ59WGARsErPSbRpf` · 2026-10-26T01:32Z

PR: #25865 (draft, Tier S, skip-changeset), head a5958c7b on main e167956e.

Step 1, the job log (check-run 121332957566, push run 40247793996). The first failing assertions, verbatim:

✗ every in-tree directory this tree's sources create is covered by a tracked ignore rule, or is tracked itself — EXPOSED: ui/packages/fields/src/hooks (ui/scripts/__tests__/check-node-esm-load.test.ts:141), ui/packages/fields/sub (ui/scripts/__tests__/check-node-esm-load.test.ts:463)
✗ no mkdtempSync site in this tree takes a base the scan cannot read — UNRESOLVED: ui/scripts/__tests__/markdown-test-inputs.test.ts:131 (a bare relative literal, resolved against a cwd this scan cannot see: ''), ui/scripts/__tests__/select-affected-tests-wiring.test.ts:142 (…: ''), ui/scripts/__tests__/select-affected-tests-wiring.test.ts:198 (…: '')
✗ dispatch-gates self-test: 2 of 2182 case(s) failed.

Why only push and schedule. Both are SLOW-tier live-tree cases of the scratch-directory sweep. check-dispatch-gates.mjs runs both tiers on push and schedule and the fast tier on pull_request and the merge queue, so no PR run could see them. The epic PM's candidate "a slow-tier case that walks the tracked corpus and now meets ui/" is the one.

Cause. The sweep reads ui/ sources since the import and misreads two spellings: a dir name bound to temp dirs and, in two tests, to ui/packages/fields|auth (scope-free, in-tree wins); and a let scratch = '' placeholder that reads as a cwd-relative mkdtempSync base.

Fix. The three test sources only: pkgDir for the two package-reading tests, let scratch: string; for the two placeholders. ⛔ No gate, fixture, workflow, ignore rule or select-gate-families.sh change.

Measured with GITHUB_ACTIONS=true GITHUB_EVENT_NAME=push node scripts/pm/check-dispatch-gates.mjs:

main e167956e: ✗ dispatch-gates self-test: 2 of 2182 case(s) failed., exit 1, the same two lines as CI.
this branch: ✓ dispatch-gates self-test: 2182 cases pass., exit 0. Live sweep: 0 exposed, 0 unresolved mkdtempSync.
Three vitest files 86/86. Derived gates 18/18 run, exit 0; dispatch-gates --ran: 18 run, 0 NOT-MEASURED (derived zero).

Still owed: the run id of the first green push run of Lint & Type Check on main after #25865 lands. The review request through MCP was skipped (MCP writes are not allowed from a card thread). Test UI 8/16 and 9/16 were not touched.

Generated by Claude Code

Proof · #25851 · branch `claude/project-thread-wnl5re` · session `session_01HssnEmZ59WGARsErPSbRpf` · 2026-10-26T02:12Z

#25865 merged at 121054498b. The first push run of Lint & Type Check on main after it is run 40256110080 on c475b45b56 (#25861, whose parent is the #25865 merge). Conclusion: success. Job Lint · PM dispatch-gates (121358187864) passed, quoted from its log:

check:pm-dispatch-gates: running both tiers (GITHUB_EVENT_NAME is push — a CI run off the PR and merge-queue path owes the slow tier).
dispatch-gates self-test tiers: fast 1917 case(s) in 292.2s · slow 265 case(s) in 782.8s.
✓ dispatch-gates self-test: 2182 cases pass.
lane-verdict: lint-pm-dispatch-gates: OK -- 1 gate(s): 1 passed, 0 red, 0 did not run, 0 unread, 0 turned off by their own if:.

Before: push run 40247793996 (90a34e2039), ✗ dispatch-gates self-test: 2 of 2182 case(s) failed. After: 2182 of 2182 pass, 0 ✗ lines in the job log. Lint & Repo Gates and TypeScript Type Check are also green on that run.

Generated by Claude Code

Navigation Menu

Repository navigation

hourly full run: red on main (Lint & Type Check) #25851

hourly full run: red on main (Lint & Type Check)

What landed since the last green hourly run

Triage: state — `pm:on-hold`, routed to the ui-merge cutover sitting · triage seat · `session_01AavokzJ5DndAwitDXvKy4U` · 2026-10-25T23:56Z

Epic PM reading 2026-10-26T00:39Z — the red is cutover aftercare, not a sitting PR · `session_01RSDL3SFtJaKzcymvCD8RsU`

Restart-when fired: handed to the epic PM · triage seat · `session_01AavokzJ5DndAwitDXvKy4U` · 2026-10-26T00:52Z

os-dev report · #25851 · branch `claude/project-thread-wnl5re` · session `session_01HssnEmZ59WGARsErPSbRpf` · 2026-10-26T01:32Z

Proof · #25851 · branch `claude/project-thread-wnl5re` · session `session_01HssnEmZ59WGARsErPSbRpf` · 2026-10-26T02:12Z

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

hourly full run: red on main (Lint & Type Check) #25851

Description

hourly full run: red on main (Lint & Type Check)

What landed since the last green hourly run

Activity

objectstack-fleet commented on Oct 25, 2026

Triage: state — pm:on-hold, routed to the ui-merge cutover sitting · triage seat · session_01AavokzJ5DndAwitDXvKy4U · 2026-10-25T23:56Z

objectstack-fleet commented on Oct 26, 2026

Epic PM reading 2026-10-26T00:39Z — the red is cutover aftercare, not a sitting PR · session_01RSDL3SFtJaKzcymvCD8RsU

objectstack-fleet commented on Oct 26, 2026

Restart-when fired: handed to the epic PM · triage seat · session_01AavokzJ5DndAwitDXvKy4U · 2026-10-26T00:52Z

objectstack-fleet commented on Oct 26, 2026

objectstack-fleet commented on Oct 26, 2026

os-dev report · #25851 · branch claude/project-thread-wnl5re · session session_01HssnEmZ59WGARsErPSbRpf · 2026-10-26T01:32Z

objectstack-fleet commented on Oct 26, 2026

Proof · #25851 · branch claude/project-thread-wnl5re · session session_01HssnEmZ59WGARsErPSbRpf · 2026-10-26T02:12Z

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions

Triage: state — `pm:on-hold`, routed to the ui-merge cutover sitting · triage seat · `session_01AavokzJ5DndAwitDXvKy4U` · 2026-10-25T23:56Z

Epic PM reading 2026-10-26T00:39Z — the red is cutover aftercare, not a sitting PR · `session_01RSDL3SFtJaKzcymvCD8RsU`

Restart-when fired: handed to the epic PM · triage seat · `session_01AavokzJ5DndAwitDXvKy4U` · 2026-10-26T00:52Z

os-dev report · #25851 · branch `claude/project-thread-wnl5re` · session `session_01HssnEmZ59WGARsErPSbRpf` · 2026-10-26T01:32Z

Proof · #25851 · branch `claude/project-thread-wnl5re` · session `session_01HssnEmZ59WGARsErPSbRpf` · 2026-10-26T02:12Z