docs: record the D-12 ruling and the Codex PR sweep; re-measure the readiness view at 42d3007f0

[Chris0Jeky]

Docs-only, three canonical files owned by the coordinator lane.

• OUTSTANDING_TASKS.md: D-12 is ticked. It records the maintainer's in-session ruling of 2026-09-05 (the Codex session's thirty-one open PRs are wanted; check, review the sensitive ones again with Opus 5, merge when ready) and the outcome: 31 fresh-context Opus 5 reviews (26 SHIP, 5 FIX-FIRST with the MEDIUMs fixed in-PR), all 31 merged as merge commits between 13:27Z and 14:32Z, three unrelated reds attributed and re-proven once, #2691 seeded, per-issue disposition comments, #1140 row 3 and #1991 L2 ticked, board moves. The section K intro no longer calls D-12 an open observation. SC-11 gains the 15:15Z re-measurement (553 remote branches, 539 without an open PR; delete_branch_on_merge still false).
• docs/releases/V0_3_0_READINESS.md: re-measured at 42d3007f0 (15:15Z). Clause 2 still 44 open; the split is 10 dogfooding / 21 ci / 13 other (#1961 closed on #2681, #2691 seeded), 21 gated / 23 un-gated, arithmetic 10 + 21 + 13 = 44 stated. Clause 4 reads in progress at the tip: the wave's last-merge run at 6cf9ef7a1 had 16 of 17 jobs green and was cancelled inside E2E Smoke when #2693 merged, which is the #2582 shape. #2691 joins the clause-4 noise list (four shapes) and the earlier "two shapes seeded, one since closed" sentence is replaced by the exact list.
• docs/STATUS.md: a coordination-lane subsection on the fifteenth block (the heading space the alpha lane left) recording the sweep: the ruling, the production changes that shipped with it, the CI noise, SC-10 unchanged (twelve parked control-plane PRs untouched), the board moves and the re-measure; the alpha lane's corrections line on the thirteenth block is left as written.

No maintainer decision is inferred: D-12 was answered directly by the maintainer in the coordinator session. Nothing else in section K or SC is ticked.

Verified: node scripts/check-docs-governance.mjs passed; git diff --check clean; every SHA, run id and count in the text was read from the live API during this session (merge SHAs from gh pr view, run ids from gh run view, the milestone split from gh issue list, branch counts from git ls-remote against the open PR list).
NOT verified: nothing executable changed; the clause-4 tip run (33974150710) was in progress when measured and is not claimed green.

Refs #2691, #2582, #2378, #2588, #1961

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, add credits to your account and enable them for code reviews in your settings.

[Chris0Jeky]

Review gate (Codex review credits exhausted, SC-9; documentation-only): one fresh-context read-only reviewer (Opus 5) on head 6009975. Verdict: SHIP, with two MEDIUMs and the cheap LOWs fixed in the fix commit; the fixes are wording and evidence, no new claims, so no second pass is owed.

Checked by reading (reviewer): every count and list in readiness section 5 closes (10 + 21 + 13 = 44; 21 gated + 23 un-gated; the ci sub-counts 15 + 2 + 1 + 1 + 2 name 21 distinct issues; the dogfooding clause names 10) and matches STATUS; clause 4 never reads green; D-12 makes no identity claim; the STATUS subsection sits inside the fifteenth block before its corrections paragraph, as the block header pre-authorizes; the only em dash added is the file's existing D-item title pattern.

Findings and disposition:

• MEDIUM, fixed: readiness line 16 said the human-gate table was re-checked at 15:15Z with no change while its SC-10 row still counted ten and named neither fix(release): fence-aware UPGRADING extraction, absolute links on the release page, and the composer suite in ci-required #2684 nor ci(release): rehearsal preview tag, workflow-revision tooling for legacy tags, and a semver changelog base #2687; the row now counts twelve and names both, and the preamble says one row changed.
• MEDIUM, fixed: the D-12 tick had no Changelog entry, which the file's rule 3 requires; the 2026-09-05 entry records the in-session directive as the rule-3 authorization.
• LOW, fixed: the section K "Nothing below is built ... until its letter is recorded" sentence now carries the D-12 exception (in effect A on the PR question).
• LOW, verified and fixed: a2fe247 is the parent of 6cf9ef7, so the sweep's last merge is one past the fifteenth block's declared bound; a range note says so.
• LOW, verified and fixed: run 33972125374 was 15 success, 1 skipped (Secret Scan), 1 cancelled, not 16 of 17; both files corrected.
• LOW, fixed: the clause-4 cell restores the cancelled-run range, the 1e234a0 run id, the 330ccb4 timestamp and the Secret Scan note; [CI][Windows] ArtefactExtractionServiceTests.ExtractAsync_ShouldPropagateCallerCancellationWithoutRecording fails intermittently on Backend Unit (windows-latest) #2691's seeding time is 13:56Z and its red is dated 03:26Z; the Frontend Unit (windows-latest) times out on slow runners: dev-up.test.mjs PowerShell spawns hit ETIMEDOUT and the job hits its timeout #2378 and [CI][dev-up] 'Node helper: TERM closes an active frontend connection' reds ubuntu Frontend Unit on a docs-only PR: frontend helper did not bind within 5 s #2588 occurrence lists gain the sweep's sightings.

Verified after the fix: node scripts/check-docs-governance.mjs passed; git diff --check clean.