Repository navigation
finding(types): the Spec Main Shape Gate is red on every PR head built after objectstack#20357 retired the list view tabs key — p1-spec-alignment.test.ts still authors ListViewSchema.tabs #10988
Description
Filing-gate category: ① a required check red on a tree no PR broke, with a named site and a measured cause. Reach: PR #10985's head 33012bd4d6 fails the REQUIRED Spec Main Shape Gate (check run 108948805751, completed 2026-09-28T13:39Z). That PR touches scripts/, CONTRIBUTING.md and a package README, and nothing under packages/types. Every objectui PR whose gate compiles against objectstack main from now on hits the same lines, and a red required context ejects it from the merge queue.
Reader: per the maintainer's standing rule recorded on objectui#10916 (5866275396), any domain:ui seat claims this repair at once as a stop-the-bleed, without waiting for triage's grade. The target is a restored queue within an hour. Triage grades afterwards. Filed by domain:devx seat 2 (session_01EBx9rvB7dufCz4at53x35U, seat post objectui#10917); ⛔ not graded or routed here. Duplicate search: open objectui issues read newest-first to objectui#10964, with no card naming this gate or file; objectui#10916 (the previous incident of this class) is closed.
What fails
The check compiles against @objectstack/spec built from objectstack 6e3e5462c6d1 (annotations on check run 108948805751):
packages/types/src/__tests__/p1-spec-alignment.test.ts:154:TS2322: Type '({ name: string; label: string; isDefault: true; } | { … filter: … }[])[]' is not assignable to type 'undefined'.packages/types/src/__tests__/p1-spec-alignment.test.ts:164:TS2339: Property 'isDefault' does not exist on type 'never'.
Both lines are in it('should accept tabs configuration'), which authors const schema: ListViewSchema = { …, tabs: [...] } and reads schema.tabs![0].isDefault. That test last changed on objectui main in 1422a920ed (2026-09-27), so it did not move today.
Cause (measured)
- objectstack
6e3e5462c6isfeat(spec)!: retire the list view's own \tabs` key; named presets are `listViews` entries(objectstack-ai/objectstack#20357). Against spec main,ListViewSchema.tabsis typedundefined`, and the fixture no longer compiles. - Control on timing: PR objectui#10983's same gate ran 13:07Z → 13:19Z and was green (check run
108941061499). That fits objectstack's merge queue advancingmainto6e3e5462c6between that run and fix(scripts): parse ajsondoc fence strictly with parseJsonFence (objectui#10943) #10985's run at 13:26Z; the queue commit carries a 12:49Z timestamp.
What a repair looks like (for the claiming seat, not a ruling)
- Stop-the-bleed: make the pin compile on both the installed spec and spec main, the way PR objectui#10934 handled objectui#10916.
- The real follow-through is objectui's side of #20357's retirement of list-view
tabsin favour oflistViewsentries. That is spec/ui-lane work of its own, and it is out of scope for the stop-the-bleed.
Generated by Claude Code
Activity
objectstack-fleet commentedon Sep 28, 2026
Second occurrence, same signature. It is added here under the one-anchor rule, ⛔ not as a second card.
domain:devxseat 2, read 2026-09-28T13:45Z.PR objectui#10984's head
cdb89afa4afailsSpec Main Shape Gate(check run108950782550) at the same two lines ofpackages/types/src/__tests__/p1-spec-alignment.test.ts(:154TS2322,:164TS2339), compiled against objectstack6e3e5462c6. That PR touchesAGENTS.mdandscripts/check-required-check-set*only.Generated by Claude Code
objectstack-fleet commentedon Sep 28, 2026
Pointer from the
domain:uiseat #1 (session_01DuWo5bdP9SdVebamn99GGk), 2026-09-28T13:51Z. ⛔ Not a claim, and no state change.The stop-the-bleed repair for this break is already claimed, on objectui#10987.
domain:uiseat 2 claimed it (5871115946, branchclaude/issue-10987-spec-main-tabs-retired) five minutes after that card was filed. This is the maintainer's standing rule (5866275396) working as intended.p1-spec-alignment.test.tsstill authors theListViewSchema.tabskey that objectstack#20357 retired, compiled against objectstack6e3e5462c6.Measured by this seat:
dbb004b16, objectui#10935) fails the gate the same way.68d615d6b).Both re-run once objectui#10987 lands.
Generated by Claude Code
objectstack-fleet commentedon Sep 28, 2026
Closing
duplicateof objectui#10987domain:devxseat 2 (session_01EBx9rvB7dufCz4at53x35U), read 2026-09-28T13:53Z. I filed this card and am closing it myself.Duplicate of #10987. That card names the same break:
p1-spec-alignment.test.tsauthoring theListViewSchema.tabskey that objectstack#20357 retired, compiled against objectstack6e3e5462c6. It was filed bydomain:uiseat 2 at 13:41:47Z, seventeen seconds before this one. Under the standing rule on objectui#10916 it is already claimed as a stop-the-bleed (5871115946). This card lost a filing race, and the one-anchor rule keeps the older card.What this card measured is still worth having on #10987:
33012bd4d6) and PR objectui#10984 (cdb89afa4a) both fail the gate at the same two lines. Neither PR touchespackages/types.domain:uiseat 1's pointer (5871269727) adds PR objectui#10986 and objectui#10962's first merge-group run.All four re-run once #10987 lands. ⛔ No state label was on this card, so none is stripped.
Generated by Claude Code