finding(changeset): a pending changeset says the spec-symbol gate skips non-exported declarations — objectui#6291 made that false, and it publishes verbatim #9528

Activity

1. objectstack-fleet

   added

   pm:dispatched

   and removed

   pm:queue

   on Sep 28, 2026
2. objectstack-fleet

   assigned

   huangyiirene

   on Sep 28, 2026
3. objectstack-fleet commented on Sep 28, 2026

   

   objectstack-fleetBot

   on Sep 28, 2026

   Last edited by objectstack-fleet

   Contributor

   More actions

   Claim: PM loop round 2
   Session: session_01EBx9rvB7dufCz4at53x35U
   Account: huangyiirene (the seat's linked user as get_me answers it; the card's assignee)
   Branch: claude/issue-9528-changeset-spec-symbol-claim
   Worktree: objectui-issue-9528
   Domain: domain:devx
   Seat: domain:devx#2
   File surface: .changeset/6286-explain-batch-cap-from-spec.md (the one sentence about scripts/check-spec-symbol-derivation.mjs), plus any other PENDING .changeset/*.md that states the same false property of that gate (stop on breach; explain in the report)
   Container & model: S, not mechanical (the corrected sentence must be re-derived from the gate's current rules), mode:subagent, model: opus (default tier; dispatch-gates.mjs --tier finds no path-derived mandate)
   Clause-②: no — release-note prose; no contract, accept set or public surface changes (a contract review is still owed: the diff is .changeset prose)
   Thread-read: none
   Serial constraints cleared: GET /pulls/{n}/files over all open objectui PRs read in this round — only the release PR objectui#5400 (changeset-release/main) touches .changeset/6286-explain-batch-cap-from-spec.md (it removes it at publication; it regenerates from main, so this fix must land BEFORE that PR merges). No in-flight claim names the file.

   Premise re-checked at claim time on objectui origin/main b120b66077 (fetched 11:40:27Z): the changeset is still pending and its paragraph still reads 「both of its scanners skip non-exported declarations」; the gate's own docblock now records module-local declarations under both rules (「rule 1 18 findings → 47 (+29 module-local declarations)」, 「rule 2 20 findings → 22 (+2 module-local declarations)」). Seven pending changesets now name the gate (the card read four).

   ---

   Generated by Claude Code
4. objectstack-fleet commented on Sep 28, 2026

   

   objectstack-fleetBot

   on Sep 28, 2026

   Contributor

   More actions

   os-dev-report
   {
   "issue": 9528,
   "status": "done",
   "branch": "claude/issue-9528-changeset-spec-symbol-claim",
   "pr": "#10970",
   "session": "session_01EBx9rvB7dufCz4at53x35U — the dispatching seat's id (subagent run; the commit trailer and PR footer carry the same id)",
   "premise_still_valid": true,
   "summary": "Rewrote one sentence in the pending .changeset/6286-explain-batch-cap-from-spec.md. It now reads 'could not catch here: when this change landed, both of its scanners skipped non-exported declarations, and this const was module-local (objectui#5899). objectui#6291 has since removed that filter from both scanners.' The frontmatter ('@object-ui/plugin-grid': patch) is unchanged. H3 is confirmed: the sentence was TRUE when it landed and went false afterwards. At c38162d (#6333, the commit that added the changeset) the gate contained 'if (!hasExportModifier(stmt)) continue;' twice, once in each scanner; 9207c53 (#7274, objectui#6291) removed both call sites. On H2, measured from code with a 2x2 probe: both scanners have no export filter today. For THIS const, however, only rule 1 (scanFile, the name match) reports it today, as a const finding with DEBT empty and no ALLOW key, so the gate would exit 1. Rule 2 (scanFileForClaims) reports it in no case, and this has nothing to do with any filter: findClaim returns null on the const's doc comment, which contains no CLAIM_PATTERNS phrase. H4: all 17 pending changesets that spell the gate were read in full, 7 by file name and 10 more by spec-symbol / check:spec-symbols. Only 6286 states the export-filter property; the other 16 are accurate on it.",
   "tests": "All runs at HEAD 25b8b6e, the only commit on the branch. (1) OS_VERIFY_LOCK_SLOT=objectui-issue-9528 os-verify-lock.sh -c 'pnpm exec vitest run --maxWorkers=2' over the 24 scripts/tests/.test.ts that git grep finds mentioning .changeset. This set is a superset of what markdown-test-inputs.mjs --changed returns for the file, which is check-changeset-no-major.test.ts; its control is plugin-dashboard/README.md, which it echoed back. Result: 'Test Files 24 passed (24)', 'Tests 914 passed (914)', 'os-verify-lock: VERDICT command-exit 0'. (2) Probe, run with no tree mutation, all under scratchpad/issue-9528/probe. Fixture A is git show c38162d^:packages/plugin-grid/src/hooks/useRecordCrudVerdicts.ts, where 'const EXPLAIN_BATCH_MAX_RECORD_IDS = 200;' is module-local. Fixture B is the same file with export added and is the control. Both were fed to the scanners of the c38162d gate (an instrumented copy that only exports scanFile; the on-disk anchor count went from 1 plain / 0 exported to 0 plain / 1 exported) and of the b120b66 gate, using today's specExportNames() (5085 names; EXPLAIN_BATCH_MAX_RECORD_IDS maps to @objectstack/spec/security). Results: old gate, fixture A: rule1 [] and rule2 []. Old gate, fixture B: rule1 const@74 and rule2 []. New gate, fixture A: rule1 const@74 and rule2 []. New gate, fixture B: rule1 const@74 and rule2 []. findClaim(doc comment) returns null. (3) Commit ancestry on a shallow checkout (is-shallow true): merge-base --is-ancestor c38162d 9207c53 exits 0; the reverse exits 1; the control leg 62311b6 (the gate's first commit, older than both) against c38162d exits 0. No ablation or reverse verification applies, because the diff is prose only.",
   "gates": [
   "pnpm check:changeset-claims: exit 0 (report-only) · 'No pending changeset names a file this change touches.' · self-contradiction corpus 1 changeset, no negated declared package · born-false 'Line addresses read in them: 0.'",
   "node scripts/check-changeset-presence.mjs: exit 0 · 'No source or published contract of a released package changed in this range, so no changeset is owed.'",
   "node scripts/check-changeset-overwrite.mjs: exit 0 (report-only) · '1 modified' · declared at base and now both '@object-ui/plugin-grid: patch' · case 2, a deliberate prose correction",
   "node scripts/check-changeset-no-major.mjs: exit 0 · 'No changeset declares a major bump.'",
   "node scripts/check-changeset-fixed.mjs: exit 0 · 'All workspace packages are in the changeset fixed group.'",
   "pnpm check:control-bytes: exit 0 · 'check-control-bytes: OK (scanned 9240 tracked text file(s); skipped 85 binary).'",
   "pnpm check:new-line-citations: exit 0 · 'VERDICT new-cross-file-line-citations: 0 new citation(s), enforcement report-only -> exit 0'",
   "node scripts/check-governed-queue-guard.mjs --test on the file: 'NOT GOVERNED'",
   "CI on PR head 25b8b6e: in_progress. One read after the PR opened showed 40 check runs: 24 success, 3 skipped, 13 in_progress, 0 failing. Convergence is the PM's to read.",
   "NOT RUN LOCALLY: pnpm check:spec-symbols, because the diff touches none of its inputs (packages//src, content/docs); pnpm lint and full pnpm test, which are CI-owned runs"
   ],
   "files_changed": [".changeset/6286-explain-batch-cap-from-spec.md (+3 -2, prose only, frontmatter unchanged)"],
   "line_budget": "not applicable: no skills/** or ledger file touched; the whole diff is +3 -2 in one file",
   "deviations": [
   "Commit trailers are the model-free pair the dispatch prescribed (Claude-Session plus 'Co-authored-by: Claude'), not the harness reminder's model-named Co-Authored-By. PR footer is the os-dev session-URL form, not the harness Generated-with form. Both follow os-dev.md and the dispatch.",
   "The test set was widened from 'every scripts/tests test that reads .changeset/' to every one that mentions it (24 files). That is a superset of the adjudicated reader."
   ],
   "mcp_calls": "0",
   "api_writes": "3 — (1) pr_create via fleet-write relay: POST /repos/objectstack-ai/objectstack/dispatches executing POST /repos/objectstack-ai/objectui/pulls, creating draft #10970; (2) label-write --assign huangyiirene via relay: POST /repos/objectstack-ai/objectstack/dispatches executing POST /repos//issues/10970/assignees, read back as MATCHES; (3) this os-dev-report comment via post-stamped: POST /repos//issues/9528/comments. There were 2 git pushes to the branch, which are not REST writes. No label written, per objectui rules; skip-changeset is never applied there.",
   "open_questions": [],
   "out_of_scope_findings": [
   "carrier: none — noted, not filed. Pending .changeset/spec-symbol-mirrors-app-shell-7265.md says 'held six declarations … Four are now bound … and three were renamed', which is 4 + 3 against six. Its commit 14a1b92 (#9465) removes six app-shell names from DEBT, so one name may have held two declarations. I did not measure this. The frontmatter is empty, so it publishes nothing and does not fall under the release-text exception. Recorded in the PR's Acceptance notes."
   ]
   }

   ---

   Generated by Claude Code
5. objectstack-fleet commented on Sep 28, 2026

   

   objectstack-fleetBot

   on Sep 28, 2026

   Contributor

   More actions

   Review — ACCEPT · PR objectui#10970 · ordinary path (not governed) · contract review PASS

   domain:devx seat 2 (session_01EBx9rvB7dufCz4at53x35U), reviewed against GitHub at 2026-09-28T12:13Z on head 25b8b6eeb2, ⛔ not against the report.
   Implemented-by: os-dev subagent of this session, branch claude/issue-9528-changeset-spec-symbol-claim · Reviewed-by: this seat · Contract review at CONTRACT_REVIEW_TIER: PASS on this head (5869591205, isolated at-tier subagent). It was owed because the diff is .changeset prose.

   Checklist

   • Shape: draft ✓ · base main ✓ · first line Fixes #9528 ✓ · the only closing keyword ✓ · Clause-②: no well-formed ✓ · assignee huangyiirene ✓.
   • Path face: .changeset/6286-explain-batch-cap-from-spec.md only, +3/−2 prose. The frontmatter ('@object-ui/plugin-grid': patch) is byte-identical to base ✓. The file is not governed.
   • The fix: the sentence was true when refactor(plugin-grid): import the batch-explain cap from @objectstack/spec/security #6333 landed and went false with objectui#6291. It now states that history in the past tense: the filter existed in both scanners at c38162d7c9, and 9207c53e45 removed it from both. It asserts no live capability of the gate, so a later gate change cannot falsify it again. The contract reviewer re-derived both facts and the commit order.
   • The dev's 2×2 probe settles H2 for this const. Today only rule 1 (the name match) would flag it. Rule 2 never would, because the const's doc comment carries no claim phrase. The export filter was what hid it from rule 1 in August.
   • H4: all 17 pending changesets naming the gate by any spelling were read. Only this one stated the export-filter property ✓.
   • Commit trailers model-free ✓. mcp_calls 0. CI on 25b8b6eeb2: 41 runs, 38 success, 3 by-design skips, 0 failures.

   Out-of-scope findings: Acceptance notes, none filed.

   • The same body's 「the minimum ^17.0.0 admits」 is dated but not false. @object-ui/plugin-grid's range has moved to ^17.4.0 (e2b38267da). 「The declared floor already carries the symbol」 still holds at release, since 17.4.0 carries it too. It stays below the finding(changeset): a pending changeset says the spec-symbol gate skips non-exported declarations — objectui#6291 made that false, and it publishes verbatim #9528 bar, which was a stated property of a gate that is now false.
   • .changeset/spec-symbol-mirrors-app-shell-7265.md counts 「six declarations」 against 4 bound + 3 renamed. The reviewer confirmed that one name, ObjectLike, sits in two places. That changeset has an empty frontmatter and publishes nothing, so under the filing gate it is not a card.

   Landing: ordinary path. It must reach main before the release PR objectui#5400 regenerates from main and merges, or the old sentence publishes. This seat flips ready and arms auto-merge in the same act.

   ---

   Generated by Claude Code
6. objectstack-fleet

   closed this as completedin #10970on Sep 28, 2026
7. objectstack-fleet

   removed

   pm:dispatched

   on Sep 28, 2026
8. akarma-synetal

   added a commit that references this issue on Oct 7, 2026

   docs(changeset): date the spec-symbol gate's export filter in the obj…

   ...

   7a9db91