Commit graph

23 commits

Author SHA1 Message Date
claude-bot-andresmgsl
44ff7524af docs(builder): re-request by head, not by verdict — a push stales every approval
The round protocol told the author to re-request exactly the reviewers
who did not approve, but the handoff predicate counts an approval only
at the current head: any push while answering a round staled the
earlier approver's verdict, doctrine said not to re-request them, and
the PR could never converge — the silent-stall shape of #26/#39.
Step 2 now re-requests by head, not by verdict: every panelist after a
push, the non-approvers alone at an unchanged head. Shape 2's wording
is aligned so the two paragraphs agree.

Defect raised by dan-claude-bot on #190; folded in at the operator's
direction while the paragraph is open. Refs #189.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-27 19:53:59 +00:00
claude-bot-andresmgsl
e9cf461ed3 docs(builder): avoid 'ledger' for review-verdict state in the red-head rule
Refs #189

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-27 18:47:00 +00:00
claude-bot-andresmgsl
203907b0ea docs(builder): green check at the head gates both request points; red-head recovery in picking
The round precondition (crew#45) at step 1 and the re-request in step 2,
with the argued exception and its evidence requirement; the ruled
classification (cancelled/stale not green, skipped/neutral green,
operator 2026-07-27); ci-red pickup precedence and crew#17's recovery
path in Picking, with the explicit shape-2 carve-out so a red head never
reads as parked.

Refs #189

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-27 18:45:03 +00:00
Daniel Marin
ede2767232
Merge pull request #176 from codex-bot-andresmgsl/build/175-post-merge-queue-state
feat: add post-merge issue queue state
2026-07-25 11:31:08 +01:00
claude-bot-andresmgsl
9ff8ed6c03 docs: a park is declared once and stands; nothing-changed resumptions post nothing
The park contract said declared-never-inferred but not that the
declaration stands, so a conservative builder re-declared every ~5
minutes — rig#145 collected 38 identical resumption audits in one
night. Now the declaration stands until the park's facts change; the
only repeat owed is the no-open-PR refresh inside the 48-hour reclaim
window. Decided on #177.

Closes #178

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-25 04:27:04 +00:00
codex-bot-andresmgsl
8e6423a07c docs: define post-merge queue state 2026-07-25 00:09:14 +00:00
claude-bot-andresmgsl
6c746364af feat: changelog_fragment_problem bounds entries at 300 characters (#167)
One definition in the fragment predicate; changelog-armed reds the PR
that writes the fragment and the assembler refuses at release, both by
inheritance. Doctrine names the number in BUILDER.md and CHANGELOG.md.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-24 18:25:46 +00:00
Daniel Marin
f168bbfbac
Merge pull request #155 from claude-bot-andresmgsl/build/154-hold-ends-on-labels
docs: a directive hold ends on the labels, and stale hold prose is triage's to correct
2026-07-24 16:10:05 +01:00
claude-bot-andresmgsl
baf0c04f1f fix: TRIAGE.md cites the real #151 needs-ruling comment; BUILDER.md timestamp to the second
The permalink for the needs-ruling ask on #151 pointed at a comment id
that does not exist; the ask is comment 5070768876 (dan-claude-bot,
2026-07-24T14:10:34Z). Also corrects the #149 claim citation from
14:11:44Z to the comment's actual 14:11:45Z.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-24 14:37:52 +00:00
claude-bot-andresmgsl
d9d73878cc docs: a directive hold ends on the labels; stale hold prose is triage's to correct
BUILDER.md shape 5 gains its ending: the hold owner's most recent
queue-label event governs over stale prose, the timeline read comes
before standing down or up on a hold, a claim against stale prose cites
the events it read, and a refused claim has two exits. TRIAGE.md
requires re-reading label events before asserting label-borne state and
makes a lifted hold's stale body header triage's to correct in the same
tick.

Closes #154

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-24 14:23:16 +00:00
claude-bot-andresmgsl
b3b9830700 docs: a post-merge acceptance criterion rides Refs #N — triage closes on the evidence
Merging #143 auto-closed #137 with a post-merge criterion unmet: the PR
carried Closes #137 as doctrine required, the criterion could only be
checked after the merge, and the contract was silent on the disagreement
between GitHub's keyword and the pipeline's close authority — the same
disagreement the cross-repo carve-out already resolved, one case over.

BUILDER.md gains the second exception beside the first: when the issue
body marks a criterion post-merge, the same-repo PR uses Refs #N and
triage closes by hand on the evidence. TRIAGE.md makes the criterion
carry its own mechanism; REVIEWER.md lists Refs #N as a spec pointer and
stops treating the reference-only PR as a defect; CONTRIBUTING.md points
at the one home instead of restating. No machinery, no label (#151 D5).

Closes #151

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-24 13:48:27 +00:00
claude-bot-andresmgsl
a3e59241b5 docs: builder surfaces write fragments, never CHANGELOG.md
CONTRIBUTING's flow item, BUILDER.md's entry rule, and the PR template
checklist all point at changelog.d/<issue>.md; the insert-above warning
retires with its anchor while the monotonic guard keeps the case (#117).
2026-07-24 11:11:23 +00:00
codex-bot-andresmgsl
17e888a6b1 docs: define directed-hold parked claims 2026-07-24 09:23:01 +00:00
claude-bot-andresmgsl
2422604f50 docs: park the handed-off claim as the fourth shape
The slot rule counted work in flight but had no shape for its most common
wait: the round passed, state:needs-human set, the human's merge pending.
Shape 4 names it, the handoff round summary is its declaration, and shape
2 now covers the round awaiting its first verdicts so the live and passed
rounds are sequential and non-overlapping (#109).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-24 07:01:07 +00:00
codex-bot-andresmgsl
8ad68192e9 docs: seed grouped changelog re-arms 2026-07-23 23:45:04 +00:00
codex-bot-andresmgsl
7791f2100e docs: define the attention contract 2026-07-23 17:52:14 +00:00
claude-bot-andresmgsl
cf2adb89f5 docs: the parked claim and the one-build-at-a-time rule
The one-issue-at-a-time bullet counted claims when it meant build work
in flight; the 2026-07-23 board (#15, #16 parked beside #73) proved the
count wrong. BUILDER.md now defines the three parked shapes, the
declared-never-inferred park comment, and unparking; TRIAGE.md names a
directed hold as a park (#77).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-23 16:37:53 +00:00
codex-bot-andresmgsl
043aeaf173 docs: define the ruling escalation contract 2026-07-23 14:57:51 +00:00
codex-bot-andresmgsl
6a9b501e05 feat: exempt offsite claims from reclaim clock 2026-07-23 12:48:04 +00:00
codex-bot-andresmgsl
5e283ec09e docs: define cross-repo discovery guards 2026-07-23 11:21:52 +00:00
claude-bot-andresmgsl
ded7f9ac04 docs: needs-ruling doctrine — LABELS row + D5–D9 rationale, escalation mechanics in TRIAGE/BUILDER/REVIEWER, changelog (#51)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-23 00:49:21 +00:00
dan-claude-bot
0d49ba4da3 docs: doctrine is a machine-verified mirror, not a pointer
Workflows are consumed by reference because GitHub materializes them at
run time; docs have no runtime — agents read the working tree — so the
agent-facing set (TRIAGE, BUILDER, REVIEWER, LABELS) vendors into each
governed repo at .ceremony/, byte-identical to the pinned ref and guarded
by docs-sync --check in CI (#19). Role files made self-contained for
vendored reading.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-22 14:22:05 +00:00
dan-claude-bot
d2aa8f1c3b docs: the agent team flow — CONTRIBUTING, LABELS, and the three role files
Discussions are where intent lives; issues are minted only by triage;
builders turn one issue into one PR; reviewers converge on verdicts;
humans decide in the discussion and at the merge. LABELS.md extends the
family taxonomy with the issue-flow labels (needs-triage, ready, claimed,
epic) the new pipeline runs on.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-22 14:13:48 +00:00