fix(runtime): a per-record action called with no recordId is refused 400 naming it, and an unknown id 404s at the door by objectstack-fleet[bot] · Pull Request #29976 · objectstack-ai/objectstack

objectstack-fleet · 2026-11-03T21:06:56Z

Requested by Yibo Wang · project thread

Fixes #29881

维护者速读

改了什么: 声明 execution: 'perRecord' 的动作（api 型或 flow 型）在两个动作门（REST /actions 与 MCP run_action）缺 recordId（或为空串）时，入流程/转发前即答 400 VALIDATION_ERROR 指名 recordId；同一 api 动作带一个调用者解析不到的 id 时，答门自己的 404 RECORD_NOT_FOUND，不再转发。
为什么改: showcase 的 showcase_recalc_estimate 是 type: 'api' 指向 flow 端点（卡片以为是 flow 型动作，实为 api 型）；缺 id 或 id 不存在时都转发进流程，拿回 400 FLOW_FAILED 和内部节点名 load_task。actions: an api action called with no recordId forwards an unresolved {id} path and answers 404 ENDPOINT_NOT_FOUND instead of naming the missing record id #28932 只按 target 里的 {id} 令牌判，这里 target 无令牌，所以改读动作自己的 execution 声明。
风险与回滚: 被拒的调用在 main 上本来就失败（同 fix(runtime): an api action whose target names its record is refused 400 naming recordId when called without one, never forwarded to a path with a hole in it #29005 的判法，Clause-②: no）；aggregate 与未声明 execution 的动作照旧无 id 运行（action door: a record-less aggregate api action that declares recordIdParam is refused 400 ("identifies its record by id"), so bulk "Recalculate Selection" never runs #29173），审批动作未声明 execution，approvals: a non-admin approver's decision from the request page answers 404 RECORD_NOT_FOUND — the action door loads sys_approval_request in the caller's data scope #28657 的转发不动。@objectstack/runtime patch。回滚 = revert 本 PR。
先看: packages/runtime/src/action-execution.ts 的 perRecordActionRecordIdRefusal 及其两处调用（buildApiActionRequest、dispatchFlowAction），和 forwardsOnDeniedSubjectLoad 新增的一行。
两分钟验证: pnpm --filter @objectstack/runtime exec vitest run src/action-flow-per-record-record-id.test.ts src/domains/api-action-forward.test.ts，预期 92 passed。
需要你做的: 无，席位评审并落地。

Clause-②: no
Metadata-protocol: none

What changes

@objectstack/runtime, action-execution.ts. A new perRecordActionRecordIdRefusal(action, recordId) answers { status: 400, message } when the action declares execution: 'perRecord' and no record id was given (undefined, null or ''). It reads only the declaration and the id. The message has the actions: an api action called with no recordId forwards an unresolved {id} path and answers 404 ENDPOINT_NOT_FOUND instead of naming the missing record id #28932 shape, and names recordId and the declaration:
```
Action 'showcase_recalc_estimate' is declared `execution: 'perRecord'` — it runs on one record — and no `recordId` was given. Pass the id of the record to run it on as `recordId`. Nothing was run.
```
It is read in two places, and both doors reach both:
- api arm: buildApiActionRequest, right beside actions: an api action called with no recordId forwards an unresolved {id} path and answers 404 ENDPOINT_NOT_FOUND instead of naming the missing record id #28932's target-token refusal, so nothing is forwarded;
- flow arm: the top of dispatchFlowAction, thrown with status: 400 and the code derived from it (VALIDATION_ERROR), so no run is created and no node's diagnostic reaches the caller.
forwardsOnDeniedSubjectLoad returns false for an action declared execution: 'perRecord'. A per-record api action whose record the caller's scope does not resolve (an unknown id, or a row it cannot read; the two are one answer by design) gets the door's shared 404 RECORD_NOT_FOUND before anything runs, as a flow or script action already did. Before, it was forwarded, and the recalc flow answered 400 FLOW_FAILED wrapping "Record … not found".
Nothing else moves. Actions declared aggregate, or declaring no execution, run record-less as before (action door: a record-less aggregate api action that declares recordIdParam is refused 400 ("identifies its record by id"), so bulk "Recalculate Selection" never runs #29173). approval_approve (the approvals: a non-admin approver's decision from the request page answers 404 RECORD_NOT_FOUND — the action door loads sys_approval_request in the caller's data scope #28657 shape) declares no execution and keeps forwarding on a denied load. action-execution.ts is not re-exported from the package entry, so the new function adds nothing to the public surface.

Why

Card #29881: POST /api/v1/actions/showcase_task/showcase_recalc_estimate {} answered 400 FLOW_FAILED with "Node 'recalc_each' failed: … Node 'load_task' failed: get_record: refusing to run — 1 filter operand(s) evaluated to nothing: recordId…". With an unknown id it answered the same code wrapping "Record … not found". The caller could not tell that it had left out an argument, and the flow's internal node names reached it.

Premise readings (measured at 18112bfa):

The card's two actions are not flow-bodied. Both are type: 'api' actions whose target (/api/v1/apps/showcase/recalc) is a type: 'flow' endpoint. actions: an api action called with no recordId forwards an unresolved {id} path and answers 404 ENDPOINT_NOT_FOUND instead of naming the missing record id #28932's check reads the target for a record token, and this target has none, so the call was forwarded with no id. The per-record signal the platform already has is the action's own execution declaration (ActionSchema.execution, An action cannot declare which bulkActions dispatch contract its body is written for — bare-string fan-out and execution: 'aggregate' deliver opposite input shapes through one authoring surface #17319). The fix reads that for the api arm, and also for type: 'flow' actions, which had the same gap.
showcase_recalc_selection is declared execution: 'aggregate'. It is NOT per-record, and action door: a record-less aggregate api action that declares recordIdParam is refused 400 ("identifies its record by id"), so bulk "Recalculate Selection" never runs #29173 deliberately lets a record-less aggregate call through, so it keeps running with no id. It is this PR's "not per-record keeps running" control. What its endpoint makes of an empty selection (the same FLOW_FAILED with node names) belongs to that flow, not the action door, and this PR leaves it as it is.
Unknown id: actions: an api action called with no recordId forwards an unresolved {id} path and answers 404 ENDPOINT_NOT_FOUND instead of naming the missing record id #28932's own arm (approval_approve) does not 404 at the door. It forwards, and the approvals route answers its own 404 REQUEST_NOT_FOUND. Matching that here would still hand the caller the flow's node names, so the per-record declaration also opts the action out of the approvals: a non-admin approver's decision from the request page answers 404 RECORD_NOT_FOUND — the action door loads sys_approval_request in the caller's data scope #28657 forward-on-denied-load exception. The door's existing 404 RECORD_NOT_FOUND envelope is used, and no new code is added.

Accept set (Clause-②: no, the #29005 reading). Every call refused now already failed on main. With no id, the forward ran a flow that refused at its first read. With an unresolvable id, it ran a flow that could not read the row as the caller (runAs: 'user'). In-repo census of execution: 'perRecord': four showcase actions. Two are operation: 'update' (that arm is untouched). Two are api actions: the recalc one, and showcase_generate_invoice_pdf → /api/v1/print/records, whose route reads the record as the caller. No plugin package declares perRecord.

Verification

Unit, packages/runtime/src/domains/api-action-forward.test.ts, new [#29881] describe, through the real HttpDispatcher REST door and invokeBusinessAction (MCP):
- with {}, { params: {} } and { recordId: '' }, the request is refused 400 VALIDATION_ERROR naming recordId, and the target is never called;
- an unresolvable id gets 404 RECORD_NOT_FOUND, and nothing is forwarded;
- controls: a real id is forwarded with { recordId }, and the aggregate twin with no id is forwarded;
- helpers: forwardsOnDeniedSubjectLoad is false for per-record and true for aggregate and undeclared; buildApiActionRequest builds no request without an id.
Unit, packages/runtime/src/action-flow-per-record-record-id.test.ts (new): a type: 'flow' per-record action, on both doors, is refused 400 naming recordId, and automation.execute is never called. An unknown id keeps the 404. A real id runs. Aggregate and undeclared flow actions run with no id.
Real HTTP, packages/qa/dogfood/test/api-action-door-forward.dogfood.test.ts, new [#29881] case on a booted stock showcase, as admin:
- showcase_recalc_estimate with {} and with { recordId: '' } answers 400 VALIDATION_ERROR naming recordId; the body contains neither load_task nor FLOW_FAILED;
- an unknown id answers 404 RECORD_NOT_FOUND with no load_task;
- a seeded task answers 200;
- showcase_recalc_selection {} is not refused at the door.
- The file has 11 tests, all passed.
Ablations, each run once and restored:
1. Unit: the three call sites removed (helper kept so the import resolves): 6 failed, every new negative pin, e.g. REST: no recordId … answers 400 VALIDATION_ERROR → AssertionError: {}: expected 200 to be 400, and the 404 pin → expected 200 to be 404. Restored: 92 passed.
2. Real HTTP: runtime rebuilt from main's action-execution.ts. The [#29881] case failed with the card's body verbatim: {"code":"FLOW_FAILED","message":"Node 'recalc_each' failed: map 'recalc_each': item 0 (subflow 'showcase_recalc_task_estimate') failed: Node 'load_task' failed: get_record: refusing to run — 1 filter operand(s) evaluated to nothing: \recordId` …"}: expected 'FLOW_FAILED' to be 'VALIDATION_ERROR'`. Rebuilt with the fix: 11 passed.
Build and suites: the @objectstack/runtime... closure build: exit 0. The @objectstack/dogfood^... closure build: exit 0. Runtime pnpm test: 540 files, 8614 passed / 19 skipped, exit 0. test:repo: 3 files, 1211 passed, exit 0. tsc --noEmit: exit 0. check:test-typecheck: exit 0.
pnpm lint: exit 0.
node scripts/pm/dispatch-gates.mjs --commands at the change set vs origin/main 18112bfa: 64 derived commands, all exit 0, with two notes:
- check-plugin-teardown-shape.mjs --self-test first exited 3 because its fixture commit 621a4876 was not in the shallow clone; after fetching it: 48 cases pass, exit 0.
- check:dual-build-cjs-loads first exited 3 because 7 package dists were missing; after building them: exit 0 (116 entry points, 71 packages).
- The 5 artifact-roster gates whose roster sits under a path this diff touches (check:authz-resolver, check:changeset-fixed, check:error-code-casing, check:filter-alias-parity, check:route-ledger-census): all exit 0.
- dispatch-gates --ran: ✓ 64 derived famil(ies) accounted for — 64 run, 0 NOT-MEASURED.
- The gates ran on commit 63068066. The pushed head 3648acbb is the same tree: only the commit message was reworded, for the trailer hook.
Serial: no open PR found touching the action door's per-record path. The one open PR a search returned (fix(core): a successful operation: 'update' action refreshes the record it wrote #29970) edits only ui/ files.

Human Minimum

1. Schemas start as Zod; packages/spec holds no runtime logic. (spec untouched)
2. Every name I added uses the casing for its kind.
3. No formatter ran; pnpm lint passes.
4. One issue, one branch off an up-to-date main, one pull request.
5. A change that publishes carries a changeset, and both declaration lines are filled in here and in the changeset.
6. packages/spec touched: n/a.
7. No hand edits to content/docs/references/, packages/*/CHANGELOG.md or content/docs/releases/.
8. Title, body and comments are in English (the 维护者速读 section is the seat's Chinese summary for the maintainer).
The Before Submitting list in CONTRIBUTING.md is done.

🤖 Generated with Claude Code

https://claude.ai/code/session_01WGufzf1rWiBfPzSuFgeCcr

…400 naming it, and an unknown id 404s at the door An action declared execution: 'perRecord' (api or flow) called with no recordId is refused 400 VALIDATION_ERROR naming it before anything is forwarded or started, through both action doors. The same api action with an id the caller's scope does not resolve answers the door's 404 RECORD_NOT_FOUND instead of being forwarded. Aggregate and undeclared actions run record-less as before (#29173). Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01WGufzf1rWiBfPzSuFgeCcr

objectstack-fleet · 2026-11-03T21:10:41Z

domain:server#6 · os-support-ai · session_01WGufzibcN9oHa8kDbTTnL5

Contract review

Served-tier: TIER_DEFAULT
Head-sha: 3648acbb91dacdc4c061dd6f330be1434725a1c0
Local-runs: none

Face: declaration (Clause-②: no, Metadata-protocol: none). No published schema and no governed path in the committed tree (5 files). Read: the full diff, card #29881 and its triage audit 6600572196, the dev report (/mnt/project-files/issue-29881/os-dev-report.md), ActionSchema.execution (packages/spec/src/ui/action.zod.ts), and #29005 (#28932's fix).

① Derived judgments

The card's mechanism was falsified, and the fix follows the measurement. Both card actions are type: 'api' forwarding to a flow endpoint whose target carries no record token, so actions: an api action called with no recordId forwards an unresolved {id} path and answers 404 ENDPOINT_NOT_FOUND instead of naming the missing record id #28932's token check never fired. The new helper perRecordActionRecordIdRefusal reads the action's own execution: 'perRecord'. It is read in both buildApiActionRequest (api arm) and dispatchFlowAction (flow arm), so REST /actions and MCP run_action both refuse 400 naming recordId before anything is forwarded or started. It reuses actions: an api action called with no recordId forwards an unresolved {id} path and answers 404 ENDPOINT_NOT_FOUND instead of naming the missing record id #28932's message shape. Correct.
The refusal matches the declared contract. The spec describes 'perRecord' as "one dispatch per selected row carrying that row's recordId", a body written for one record, and says to omit execution only when the body serves both. A call with no id is outside that contract. aggregate and undeclared actions keep running record-less (action door: a record-less aggregate api action that declares recordIdParam is refused 400 ("identifies its record by id"), so bulk "Recalculate Selection" never runs #29173 pinned by recalc_selection). The in-repo census is 4 showcase actions, and each refused call already failed on main. Clause-②: no follows the fix(runtime): an api action whose target names its record is refused 400 naming recordId when called without one, never forwarded to a path with a hole in it #29005 reading. Accepted.
Unknown id: forwardsOnDeniedSubjectLoad now returns false for a perRecord action, so an id the caller's scope does not resolve gets the door's own 404 RECORD_NOT_FOUND (one answer for unknown and unreadable, by design) instead of a forwarded flow failure naming load_task. approval_approve declares no execution, so the approvals: a non-admin approver's decision from the request page answers 404 RECORD_NOT_FOUND — the action door loads sys_approval_request in the caller's data scope #28657 forward is unchanged. Correct; nothing internal leaks on either path.
Pins: unit (api arm + flow arm, empty id is no id), runtime forward tests, and the dogfood door test (both card actions: no id → 400 naming recordId with no node name; unknown id → 404; aggregate runs). Ablations went red (6 unit failures; dogfood returned the card's exact FLOW_FAILED body) and were restored.

② Semver level

'@objectstack/runtime': patch. perRecordActionRecordIdRefusal is exported from action-execution.ts; the dev reports no entry-export change, so the package surface does not move. Consistent with both declaration lines.

③ Boundary flags

No open_questions. An out-of-repo action declared perRecord whose body happens not to read its record, called with no id, is now refused where it ran before. That is the declared contract being enforced (an author whose body serves both omits execution). The changeset names the rule. Not filed.

Implemented-by: claude/project-thread-mtl298
Reviewed-by: session_01WGufzibcN9oHa8kDbTTnL5

VERDICT: PASS

objectstack-fleet · 2026-11-03T21:12:56Z

domain:server#6 · os-support-ai · session_01WGufzibcN9oHa8kDbTTnL5

席位核对

Verdict: ACCEPT at head 3648acbb91dacdc4c061dd6f330be1434725a1c0 · 2026-11-03T21:10Z

Shape: 5 files: a runtime action-door helper read by the api and flow arms, the denied-load forward exemption, unit and forward tests, the dogfood door test, and one patch changeset. No published schema, no governed path ⇒ not Tier H. TIER_DEFAULT review PASS 6615520067 at this head.
Premise measured: the dev reproduced the card's FLOW_FAILED body through the dogfood door test (ablation). The mechanism differs from the card: these are api actions forwarding to a flow, so the fix reads the action's execution: 'perRecord'.
Gates: dispatch-gates 64/64 exit 0, --ran 0 NOT-MEASURED. Runtime unit, forward and dogfood suites pass. Ablations went red and were restored. Landing waits for every check on this head to be green or an expected skip.
Accepted deviations: the harness-pinned branch claude/project-thread-mtl298. The landing site moved from the token check to the execution declaration, measured. The signed trailer is appended by the seat through the relay.
Landing: on all-green at this head, the seat sends pr_ready + automerge_enable through the relay.

github-actions · 2026-11-03T21:16:21Z

📓 Docs Drift Check

This PR changes 1 package(s): @objectstack/runtime, touching 5 documentable anchor(s).

2 hand-written doc(s) NAME something this change touched and may need an implementation-accuracy re-verification:

content/docs/ui/actions.mdx (via dispatchFlowAction (symbol, a top-level function), perRecord (literal, a string literal in a comment on a changed line; a string literal in forwardsOnDeniedSubjectLoad; a string literal in perRecordActionRecordIdRefusal))
content/docs/ui/views.mdx (via perRecord (literal, a string literal in a comment on a changed line; a string literal in forwardsOnDeniedSubjectLoad; a string literal in perRecordActionRecordIdRefusal))

⛔ 1 release-owned page(s) also name something this change touched. These are read-only:

content/docs/releases/v17/17-5.mdx (via perRecord (literal, a string literal in a comment on a changed line; a string literal in forwardsOnDeniedSubjectLoad; a string literal in perRecordActionRecordIdRefusal))

content/docs/releases/ is RELEASE-OWNED (AGENTS.md "Documentation Guardrails"): release
notes are written centrally at release time, and a code PR that edits them is the exact PR
that guardrail exists to stop. They are still audited — read-only. If one of them is actually
wrong, file an issue or open a dedicated docs-only PR; do not edit it here.

What this run could not see

the SDK route bridge reached 61 of 219 client-bound route-ledger rows — the other 158 have no registrar path: tail to select them, so pages documenting THEIR client methods cannot appear above, on this or any run. Of those 158: 0 are remediable by widening that discovery convention (an in-repo file declares the path; the convention did not scan it); 55 are structural — on a ledger where NOT ONE row is declared in-repo, so no discovery change reaches them at any price; 103 are undecided (no in-repo declaration, on a ledger that has other in-repo registrars — absence and an unreadable spelling are not distinguishable here). The rows themselves: node scripts/docs-audit/affected-docs.mjs --bridge-coverage
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 — 28 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 18112bfa776b896f705c182f07b8baeff45957a7 → packageMentionDocs.

Which tree this was computed on

This run read content/docs from 413c316ccfcc5c8471f07525875e78c6af95d651 — the merge of head e2aec5f47e559fbdcff590526fe1c0414d809f94 into base 18112bfa776b896f705c182f07b8baeff45957a7, which is what actions/checkout gives a pull_request run. Not the PR head.

A worktree cut from an older main holds a different content/docs, so re-deriving there can legitimately return a different list — that is a different tree, not a wrong row. To answer on the same tree:

# while this PR is open — GitHub drops the merge commit once it closes
git fetch origin 413c316ccfcc5c8471f07525875e78c6af95d651 && git checkout 413c316ccfcc5c8471f07525875e78c6af95d651
# afterwards, rebuild it from the two parents, which stay fetchable
git fetch origin 18112bfa776b896f705c182f07b8baeff45957a7 e2aec5f47e559fbdcff590526fe1c0414d809f94 && git checkout -B drift-repro 18112bfa776b896f705c182f07b8baeff45957a7 && git merge --no-ff e2aec5f47e559fbdcff590526fe1c0414d809f94

node scripts/docs-audit/affected-docs.mjs --json 18112bfa776b896f705c182f07b8baeff45957a7

⚠️ That checkout carried uncommitted changes, so the commit above does not fully identify what was read.

Advisory only, and a precision-first one (#9192): a page is listed because it names a
symbol, wire route or SDK method this diff touched — not because it mentions a changed
package. Each row says which anchor put it there, so a wrong row is reportable rather than
merely annoying. To re-verify, run the docs-accuracy-audit workflow scoped to these files:
node scripts/docs-audit/affected-docs.mjs 18112bfa776b896f705c182f07b8baeff45957a7 → pass the list as
args.docs, on the commit named under Which tree this was computed on.

…, not run record-less Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01WGufzf1rWiBfPzSuFgeCcr

objectstack-fleet · 2026-11-03T21:23:24Z

domain:server#6 · os-support-ai · session_01WGufzibcN9oHa8kDbTTnL5

席位核对

Verdict: ACCEPT at head e2aec5f47e559fbdcff590526fe1c0414d809f94 · 2026-11-03T21:22Z

Supersedes ACCEPT 6615564413 at 3648acbb. The head moved by one dev docs commit, prose only (head-move exception: prose surface, original record carries on): content/docs/ui/actions.mdx gains one sentence saying a perRecord api or flow action with no recordId is refused 400 VALIDATION_ERROR and an unresolved record answers 404 RECORD_NOT_FOUND. That matches the behaviour judged in review PASS 6615520067. No code, test or changeset moved. The 26 docs commands dispatch-gates derived for the new change set exit 0.
Landing: on all-green at this head, the seat sends pr_ready + automerge_enable through the relay.

github-actions Bot added size/m documentation Improvements or additions to documentation tests tooling labels Nov 3, 2026

docs(actions): a per-record action called with no recordId is refused…

e2aec5f

…, not run record-less Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01WGufzf1rWiBfPzSuFgeCcr

objectstack-fleet Bot marked this pull request as ready for review November 3, 2026 21:59

objectstack-fleet Bot added this pull request to the merge queue Nov 3, 2026

Merged via the queue into main with commit 30dcd8a Nov 3, 2026
50 of 51 checks passed

objectstack-fleet Bot deleted the claude/project-thread-mtl298 branch November 3, 2026 22:53

objectstack-fleet Bot mentioned this pull request Nov 3, 2026

Actions: a per-record flow action called with no recordId runs the flow and answers 400 FLOW_FAILED naming an internal node, instead of refusing at the action door with the missing recordId named #29881

Closed

github-actions Bot mentioned this pull request Nov 4, 2026

4-hourly full run: red on main (CI) #29494

Closed

objectstack-fleet Bot mentioned this pull request Nov 4, 2026

feat(spec)!: retire emailAndPassword.revokeSessionsOnPasswordReset — a completed password reset always ends every session the account holds #29964

Merged

Navigation Menu

Repository navigation

fix(runtime): a per-record action called with no recordId is refused 400 naming it, and an unknown id 404s at the door - #29976

fix(runtime): a per-record action called with no recordId is refused 400 naming it, and an unknown id 404s at the door#29976
objectstack-fleet[bot] merged 2 commits into
mainfrom
claude/project-thread-mtl298

objectstack-fleet Bot commented Nov 3, 2026 •

edited

Loading

Uh oh!

objectstack-fleet Bot commented Nov 3, 2026

Uh oh!

objectstack-fleet Bot commented Nov 3, 2026

Uh oh!

github-actions Bot commented Nov 3, 2026 •

edited

Loading

Uh oh!

objectstack-fleet Bot commented Nov 3, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

objectstack-fleet Bot commented Nov 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

维护者速读

What changes

Why

Verification

Human Minimum

Uh oh!

objectstack-fleet Bot commented Nov 3, 2026

Contract review

① Derived judgments

② Semver level

③ Boundary flags

Uh oh!

objectstack-fleet Bot commented Nov 3, 2026

席位核对

Uh oh!

github-actions Bot commented Nov 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

📓 Docs Drift Check

Uh oh!

objectstack-fleet Bot commented Nov 3, 2026

席位核对

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

objectstack-fleet Bot commented Nov 3, 2026 •

edited

Loading

github-actions Bot commented Nov 3, 2026 •

edited

Loading