fix(#4670): bound the commit-claim window to the plan's own history (#4813)

* test(#4670): add failing-first coverage for the bounded commit-claim window

* fix(#4670): bound the commit-claim window to the plan's own history

The reconciliation measured plan_head_before..HEAD — a window that grows
with every later plan's task and SUMMARY commits plus execute-phase's own
phase-completion commit — so an honest plan flagged commit_claim_mismatch
as soon as anything landed after it (real project: claims 3/5/2 measured
20/10/5).

The executor now also records plan_head_after (HEAD at its measurement
moment, after the last task commit, before the SUMMARY commit), and
verify-work reconciles exactly against plan_head_before..plan_head_after
with a merge-base ancestry check; SUMMARYs without the anchor fall back to
the legacy warning path instead of an unsound BLOCKER. Both #3968 failure
modes (claimed commits never made; task commits lost) still block, driven
by the issue's own fixture scenarios.

Emitted-Drift-Ack-Growth: verify-work.md — reconciliation gains the bounded window and legacy fallback (#4670)
Emitted-Drift-Ack-Growth: gsd-executor.md — plan_head_after anchor documented in the measurement protocol (#4670)

* fix(#4670): name the history-rewrite case and keep the executor under its cap

Review findings: the BLOCKER enumeration named only the two #3968 causes,
so an honest plan whose recorded window was rewritten afterwards (rebase,
amend, cherry-pick) got a mislabeled diagnosis — the clause now names that
case with the manual-recount remedy. The executor's growth crossed the
LARGE-tier hard cap (49152), so the plan_head_after documentation is
compressed to the minimal capture + frontmatter write (verify-work.md
carries the semantics), the #2751 PROSE_ALLOWLIST entry is re-pointed at
the shifted line (#4670 moved it from 823 to 825), the compact-content
baseline is regenerated, and the changeset records the two un-established
edges (shared-base waves, subrepo ledgers).

Emitted-Drift-Ack-Growth: gsd-executor.md — plan_head_after anchor documented in the measurement protocol (#4670)

* docs(#4670): backfill changeset PR number

* docs(#4670): backfill changeset PR number

---------

Co-authored-by: sim <sim@local>
This commit is contained in:
Tom Boucher
2026-09-16 22:06:03 -04:00
committed by GitHub
parent 652796e903
commit 651511d1e3
6 changed files with 108 additions and 20 deletions

View File

@@ -43,8 +43,16 @@ describe('#3968 — measured commit claims', () => {
'verify-work must run a commit-claim reconciliation over each SUMMARY');
assert.ok(verify.includes('ACTUAL=$(git rev-list --count "${BASE}"..HEAD)'),
'the reconciliation uses the SAME instrument as the executor (rev-list over the recorded base)');
assert.ok(/ACTUAL == CLAIMED \+ 1/.test(verify),
'the post-measurement SUMMARY commit is an expected +1, not a false BLOCKER');
// #4670: the +1 tolerance is retired for bounded SUMMARYs (exact equality
// over plan_head_before..plan_head_after); the literal survives only in
// the legacy-fallback retirement sentence. Pin the shipped rule, not the
// retired tolerance:
assert.match(verify, /Bounded reconciliation \(#4670\)/,
'the bounded exact-equality reconciliation is the shipped rule');
const reconciliationIdx = verify.indexOf('Commit-claim reconciliation');
const legacySentenceIdx = verify.indexOf('`ACTUAL == CLAIMED + 1` tolerance was a guess');
assert.ok(legacySentenceIdx > reconciliationIdx,
'the +1 tolerance appears only as the retired legacy rule inside the reconciliation block');
assert.ok(/BLOCKER/.test(verify.slice(verify.indexOf('Commit-claim reconciliation'), verify.indexOf('Commit-claim reconciliation') + 1800)),
'a mismatch must be flagged BLOCKER — the phase must not read as done');
});
@@ -59,3 +67,61 @@ describe('#3968 — measured commit claims', () => {
'the uncommitted_files field is defined by the porcelain command, not a narrated []');
});
});
// ── #4670 — the commit-claim window is bounded to the plan's own history ─────
// `plan_head_before..HEAD` grows with every LATER plan's commits and
// execute-phase's phase-completion commit, so an honest plan flagged as
// `commit_claim_mismatch` BLOCKER as soon as anything landed after it. The
// executor now also records `plan_head_after:` (HEAD at its measurement
// moment — after the last task commit, before the SUMMARY commit), and
// verify-work reconciles against that bounded window with EXACT equality;
// legacy SUMMARYs without the anchor fall back to the WARNING path.
describe('#4670 — bounded commit-claim reconciliation', () => {
test('executor records plan_head_after — the plan window bound (#4670)', () => {
const executor = read('agents/gsd-executor.md');
assert.ok(
executor.includes('plan_head_after: ${PLAN_HEAD_AFTER}'),
'the SUMMARY frontmatter must carry plan_head_after, captured at the measurement moment'
);
assert.ok(
/PLAN_HEAD_AFTER=\$\(git rev-parse HEAD\)/.test(executor),
'PLAN_HEAD_AFTER must be captured from HEAD at the same measurement moment as the count'
);
});
test('verify-work bounds the commit-claim window to the plan own history (#4670)', () => {
const verify = read('gsd-core/workflows/verify-work.md');
const afterIdx = verify.indexOf('AFTER=$(grep -oE \'^plan_head_after: [0-9a-f]{7,40}\' "$SUMMARY_FILE" | awk \'{print $2}\')');
assert.ok(afterIdx !== -1, 'the reconciliation must extract plan_head_after');
assert.ok(
verify.includes('git merge-base --is-ancestor "$AFTER" HEAD'),
'the plan end must be verified to still be in history (ancestor check)'
);
assert.ok(
verify.includes('git rev-list --count "${BASE}..${AFTER}"'),
'the measured count must be bounded to the plan own window (BASE..AFTER)'
);
});
test('bounded mismatch stays a BLOCKER — #3968 failures still caught (#4670)', () => {
const verify = read('gsd-core/workflows/verify-work.md');
const boundedIdx = verify.indexOf('#4670');
assert.ok(boundedIdx !== -1, 'the bounded reconciliation section must exist');
const section = verify.slice(boundedIdx, boundedIdx + 2400);
assert.match(section, /BLOCKER/, 'a bounded mismatch must remain a BLOCKER');
assert.match(section, /commit_claim_mismatch/, 'the mismatch verdict name is kept');
});
test('legacy SUMMARYs without plan_head_after fall back to the warning path (#4670)', () => {
const verify = read('gsd-core/workflows/verify-work.md');
const afterNeedle = 'plan_head_after';
assert.ok(verify.includes(afterNeedle), 'plan_head_after must appear in verify-work');
// The legacy clause: a base without the anchor cannot be judged by the
// unsound unbounded window — it is a WARNING with the measured state.
assert.match(
verify, /WARNING[\s\S]{0,400}plan_head_after/,
'SUMMARYs without plan_head_after must fall back to the WARNING path, not the unsound BLOCKER'
);
});
});