[garrytan]

{
  "schema_version": 1,
  "pr": 3567,
  "head": "621baec9b8e50a74f7475b98860289ba354f32df",
  "version": "1.91.299.0",
  "base": "gsta147-carry-rules",
  "gate": {
    "remote_gate": {
      "pass": 23668,
      "fail": 0,
      "skip": 0
    }
  },
  "session_kind": "unattended",
  "pregate": "requires-remote cleared @2b975190e0b1 (1 checks, current for 621baec9b8e5)",
  "requires_remote": [
    "free-tests.yml/cso-macos-launcher: https://github.com/garrytan/gstack/actions/runs/40803874016",
    "free-tests.yml/macos-named-regressions: https://github.com/garrytan/gstack/actions/runs/40803874016",
    "windows-free-tests.yml/windows-free-shard: https://github.com/garrytan/gstack/actions/runs/40803877944",
    "pregate.json/remote_gate: remote-gate tree=2b975190e0b1 tests=23668"
  ],
  "issues": {
    "fixes": [
      3559
    ],
    "refs": [],
    "flakes": []
  },
  "docs": {
    "status": "updated",
    "scope": "docs/, README.md, CONTRIBUTING.md, AGENTS.md, CLAUDE.md against 3b90122..45d9835 (review depth, receipt depth fields, queue checks, diff-scope --head)"
  },
  "review_loop": {
    "passes": 2,
    "blocking_found": 2,
    "deferred": 0,
    "noted": 9,
    "fixed": 7,
    "scope": "frozen",
    "dispatches": {
      "specialist": 7,
      "red-team": 2
    }
  }
}

Stacked on #3564 (receipt carry rules). Review that one first; this PR's diff is against its branch.

What changed

A 12-line code fix now gets light depth: one batched specialist plus the native adversarial pass, and the plan prints the depth and gate scan the receipt records.

Before/after as text: A 12-line code fix now gets light depth: one batched specialist plus the native adversarial pass, and the plan prints the depth and gate scan the receipt records

--- Before · before.txt · captured output
+++ After · after.txt · captured output
 $ gstack-review-plan --base main   # a 12-line fix to lib/util.js
 REVIEW_CLASS: code (code: lib/util.js)
+REVIEW_DEPTH: light runtime_lines=12 rules=9bc0e4d2f74ef6b4 reasons=runtime=12<=50 codes=none
+GATE_SCAN: tagged=0 head=<HEAD>
+REVIEW_PLAN: class=code pass=1 next=full dispatches=2 numbers=no fingerprint=none
-REVIEW_PLAN: class=code pass=1 next=full dispatches=3 numbers=no fingerprint=none
 NEXT: full (pass 1 reviews the whole diff)
+DISPATCH: specialist batched effort=full checklists=testing,maintainability,security,performance,state-machines,red-team
+DIFF_CMD: batched git diff <diff-base>
+DISPATCH: adversarial-native adversarial effort=full reason=backtest inconclusive (GSTA-147 D5)
+DIFF_CMD: adversarial git diff <diff-base>
+SKIP: red-team (light depth: the batched specialist carries the Red Team prompt)
+SKIP: outside-voice (code class, no numeric claims)
-DISPATCH: specialist state-machines effort=full
-DISPATCH: adversarial-native adversarial effort=full
-DISPATCH: outside-voice outside-voice effort=host
-SKIP: red-team (code class, 12 lines, no critical)
 

• Review depth for /ship (lib/review-depth.ts, new lib/review-collect.ts, bin/gstack-review-plan, the review-army and adversarial sections). The depth depends on runtime lines: code-class lines minus docs, tests, and outputs registered with a command in the base commit's .gstack/generated.json.
  • light (50 or fewer runtime lines): one batched specialist (testing, maintainability, security, plus any scope-selected checklists, with the Red Team prompt) and the native adversarial pass. At 0 runtime lines it runs the core checklist only.
  • standard (more than 50): the class's specialists, each scoped to its hunks with DIFF_CMD pathspecs.
  • full: gate class or any classifier error, unchanged.
  • The outside voice runs whenever the diff publishes numbers. A fixed or open blocker, or a CRITICAL, raises light to standard for the rest of the invocation, and --depth-floor only raises. Every DEPTH_ERROR: carries its input, the sanitized cause and the repair.
• The receipt records the depth. write adds gstack_version and review_depth: depth, class, runtime lines, rules hash, base and registry commits, the gate scan and reasons. They are computed at the checked-out head from the committed range, and --head must be HEAD; tracked edits in the checkout make the gate scan unavailable, which records full. The SHIP_RECEIPT: line prints depth=<d> after artifacts_consumed.
• The queue enforces it. ready and train build recompute the depth the head needs from the same committed range and refuse a lower receipt with QUEUE_RECEIPT_DEPTH_RAISED. The REVIEW_DEPTH: line prints the receipt's depth, the required depth, runtime lines, class, reasons and inputs. A gate scan counts only when it names the full bound commit; otherwise the head needs full.
• A carry recomputes the depth. It rebinds the gate scan to the new head and refuses when the new head needs more (a test change that skips a test makes the diff gate class), when --head is not the checked-out HEAD, when the receipt records gstack_version without review_depth, or when an owner exception no longer stands.
• review_loop_exception (write --review-loop-exception) records an accepted stop after a fix-check edit under frozen scope: who, when, via, why, and the fix-check's head and findings. It needs review_loop.fix_checks of 1 or more and a fix-check head that is the receipt head or its ancestor, and the body must carry the printed line word for word (REVIEW_LOOP_EXCEPTION_INVALID). The queue checks this exception and the docs exception the same way, including the fix-check ancestry (QUEUE_RECEIPT_EXCEPTION_INVALID).
• gstack-diff-scope <base> --head <sha> reads the committed range only, the scope the writer and the queue share.
• .gstack/ship-policy.json min_gstack_version goes from 1.91.146.0 to 1.91.299.0.

Fixes #3559

Why

"A 12-line fix paid for the same reviewer fan-out as a change to the merge gate." The plan (GSTA-147, approved by Garry: "rec" on gate_rev 2) sizes the review to what the diff can break, while gate code, security paths, published numbers and every classifier failure keep the full review.

Light-depth backtest (D5): INCONCLUSIVE, so light keeps the adversarial pass.

Repo	Light-eligible PRs	Attributable findings	Blockers a light review would have lost	Inconclusive
gstack	7	3 (#3542, #3517, #3503)	0	4 (#3549 had 2 CRITICALs, one deferred as #3548; #3545, #3532, #3530 had no attributable reviewer roles)
gstack-evals	2	2 (#17, #2)	0	0

The bar was at least 15 light-eligible PRs with attributable findings and 0 blockers lost. Only 5 were attributable, so light at runtime > 0 dispatches the native adversarial pass beside the batched specialist (reason "backtest inconclusive (GSTA-147 D5)").

Blast radius

The change touches /ship's review dispatch, every receipt write and the queue's acceptance of receipts. Depth only lowers the fan-out on non-gate diffs of 50 or fewer runtime lines, and receipts with neither new field (every receipt written before this release) are accepted as legacy, so in-flight PRs keep moving (script ran: test/ship-receipt-depth.test.ts covers the legacy case, and the whole free tier passed on the frozen head). Each write runs gstack-diff-scope and the gate scan once, about 0.4 s.

These resolutions differ from the plan:

• The gate scan lives in review_depth.gate_scan, not review_loop.gate_scan, because review_loop comes from the review-scope summary.
• write computes the depth itself. Step 9 plans on the uncommitted tree before Step 15 commits, so its scan names an older head.
• A carry's depth refusal uses RECEIPT_CARRY_REFUSED, not a new RECEIPT_CARRY_DEPTH_RAISED code.
• min_gstack_version rises to the newest released version, not to this PR's own version. That version is allocated at the slot stamp, and the policy file is not a stamp path. A one-line follow-up after the stamp can raise it.
• ready has no --json mode, so the diagnostics are text only.

Approvals

Date	Asked	Exact words	Outcome	Covers	Still pending
2026-10-28	Approve GSTA-147 plan v2 (proportional review depth) and the 12 open-highs resolutions	“rec” (Garry (via GStack Minder))	approved	plan v2 at gate_rev 2, every recommendation	none
2026-10-28	Approve the #3559 amendment: a structured review_loop_exception on the receipt	“It only tightens what's already recorded in prose, so it doesn't need another round.” (GStack Minder (relaying Garry's plan approval))	approved	the review_loop_exception field and its queue check	none

Source: gstack-approvals record for gsta147-review-depth (display only; not an approval citation).

The #3559 amendment came from GStack Minder's relay of Garry's plan approval; the lane never merges.

Summary

Proportional review depth for /ship (light, standard, full), receipts that record the depth bound to the head they describe, and a queue that refuses a receipt below what the head needs. The pre-landing review tightened that binding, so the queue and the writer now read the same committed range, a gate scan binds only to the full commit, and carry refuses the cases the queue would refuse.

Test Coverage

Step 7 audit (audit-only): 25 of 28 paths tested.
Tests: 92 → 124 (+32 new) across the seven touched test files.
Coverage: 89% of paths tested (86% at two stars or better), with 3 gaps and 1 weakly covered path.
Test value: the audit-only run recorded no per-card counts; 1 path weakly covered.

Gaps left as proposed tests: a test/ship-queue-train.test.ts row that builds a train with a light receipt on a standard head; the rules-only equal-tree acceptance; a bounded recompute time for a five-PR train (filed in TODOS.md).

Tests ran: remote_gate 23668 (pregate, whole free tier on one Ubicloud VM)

Pre-Landing Review

Pass 1 (full, gate class: 6 specialists, Red Team, native adversarial, outside voice on openai/gpt-6.1-sol) found 14: 7 fixed in 4ad8a6b, including 2 blocking (the queue bound a gate scan by prefix and recomputed scope without the writer's range) and a reproduced CRITICAL (a carry --head other than the checkout recorded a working-tree gate scan). Pass 2 on the fix range (one batched specialist over six checklists plus Red Team) found 2 notes and nothing at the floor.

REVIEW_LOOP: passes=2 blocking_found=2 deferred=0 noted=9 fixed=7 scope=frozen dispatches=9

Gate edits: 19 listed, 14 read, 0 findings — 0 kept, 0 restored, 0 open

Outside review: completed (openai/gpt-6.1-sol, pass 1); its carry finding matched the testing specialist's and was fixed.

Review notes

• bin/gstack-ship-receipt:603 (state-machines, pass 2): after a carry across rewritten history, a later linear carry checks the old fix-check ancestry and refuses a receipt the queue accepts. Fails closed (a needless /ship rerun).
• bin/gstack-diff-scope:87 (simplification, advisory): the first clause of the --head test is redundant.
• bin/gstack-review-plan:120 (security): control characters in a path can reach DIFF_CMD.
• lib/review-collect.ts:129 (adversarial): collectLive follows a FIFO or symlink in the working tree.
• lib/ship-receipt-depth.ts:121 (adversarial): legacy acceptance has no expiry.
• bin/gstack-ship-receipt:208 (Red Team): review_depth is the depth the head needs, recomputed at write, not the depth the review ran.
• Advisory: deeper() is dead at bin/gstack-ship-receipt:605; Collected.mergeBase is unused; the depthOfCommitInput floor is unused.

Exploratory QA

Not run as a separate charter: the change is CLI and library behaviour, exercised through the fixture repositories in test/ship-receipt-depth.test.ts, test/review-plan-cli.test.ts and test/diff-scope.test.ts, plus the before/after review plan above, captured on a scratch repository.

Design Review

No frontend files changed — design review skipped.

Eval Results

Not run locally. eval:select falls back to the full PR profile (143 behaviors) because test/size-ledger/ship/gsta147-review-depth.json is an unknown dependency, beyond the lane's $10 focused cap; the paid coverage for this PR is the train gate (gstack/full-gate), still outstanding.

Greptile Review

No Greptile comments.

Plan Completion

Step 8 audit: 21 of 33 DONE, 7 CHANGED, 3 PARTIAL, 2 NOT DONE (plan: captain-gstack-minder/plans/GSTA-147/plan.md, PR 2).

• Deferred (P1 in TODOS.md): carry.covers derived from the depth (until then a carried light receipt's covers is the fixed list); a bounded recompute-time test for a five-PR train.
• PARTIAL, not fixed: QUEUE_RECEIPT_DEPTH_RAISED prints neither both rule hashes nor runtime lines at both heads; the rules-only equal-tree acceptance with a note is not implemented and has no test.

Issues

Fixes #3559

Verification Results

Plan obligations run at review: the depth matrix, fail-closed errors, write, carry and queue refusals and the review-loop exception PASS in their focused tests; the five-PR recompute bound is NOT RUN (deferred above).

TODOS

No TODO items completed in this PR. Added a P1 section for the two deferred plan items.

Documentation

Status: updated. Factual edits to docs/ship-policy.md and docs/troubleshooting.md; no blockers.

Scope: the committed range 3b90122..45d9835 (40 files; nothing staged, unstaged or untracked): review depth (lib/review-depth.ts, lib/review-collect.ts, bin/gstack-review-plan), receipt fields gstack_version, review_depth and review_loop_exception (bin/gstack-ship-receipt, lib/ship-receipt-depth.ts), the queue checks QUEUE_RECEIPT_DEPTH_RAISED and QUEUE_RECEIPT_EXCEPTION_INVALID, and gstack-diff-scope <base> --head <sha>. Docs that name these commands were audited.

Documentation health:

• docs/review-policy.md: Current (depth table, runtime lines, fail-closed list and gate paths match lib/review-depth.ts)
• docs/ship-policy.md: Updated (shared-queue ready, /ship-before-READY and train build refusal lists name the two new queue codes; verify --train narrowed to QUEUE_RECEIPT_MISSING/STALE, since it does not recompute depth; the carry depth and exception refusals name RECEIPT_CARRY_REFUSED)
• docs/troubleshooting.md: Updated (RECEIPT_INVALID covers write --head not being the checked-out HEAD; QUEUE_RECEIPT_DEPTH_RAISED names the base ref per command; QUEUE_RECEIPT_EXCEPTION_INVALID adds the fix-check ancestor rule; RECEIPT_CARRY_REFUSED adds the depth and owner-exception refusals)
• CHANGELOG.md: Current (not edited; the parent owns it)
• TODOS.md: Current (not edited)
• docs/pregate.md, docs/captain.md, docs/unattended.md, docs/model-policy.md: Current
• README.md, CONTRIBUTING.md, AGENTS.md, CLAUDE.md, docs/skills.md: Current (none names the changed commands or codes)

Documentation Debt

• ⚠️ gstack-diff-scope <base> --head <sha>: reference only (bin header, ship-policy.md, CHANGELOG); no how-to.
• ⚠️ .gstack/generated.json as a review-depth input: explained in review-policy.md, but pregate.md's generated-v1 section does not point to it (reference cross-link).

Diagram drift: none (no architecture diagram names the review or receipt machinery).

Test plan

• gstack-pregate --stage tests --remote --spend-cap 2 --timeout 2700 --allow-repo-commands: PREGATE_REMOTE tree=2b975190e0b1 tests=23668 failed=0 spent=0.08; PREGATE_VERIFY ok (macOS and Windows lanes cleared)
• Free Tests (macOS lanes included) and Windows Free Tests dispatched at the head: both green at 621baec (https://github.com/garrytan/gstack/actions/runs/40803874016, https://github.com/garrytan/gstack/actions/runs/40803877944)
• bun test test/approval-classes.test.ts test/gstack-ship-receipt.test.ts test/ship-receipt-depth.test.ts: 37 pass, 0 fail for the first two after the summary-line fix; ship-receipt-depth 6 pass
• bun run typecheck: clean

🤖 Generated with Claude Code

[github-actions]

E2E Evals: ✅ PASS (flaky: 3 case(s) passed on their one rerun)

[test:paid] VERDICT GREEN — lane gate/pr, attempt 1
  rule 112/112 · behavior 1/1 · judge 24/24 · quarantined 0 (0 failing the lane)
  SKIPPED 0 · INFRA 0 · INCOMPLETE 0 · unattributed 0 · ACTION REQUIRED 0
  wall 6m02s · cost $32.74 known + 32 of 139 trial(s) cost unknown
  FLAKY 3 (failed once, passed on its one rule-16 rerun) · 3 rerun execution(s), cost $2.75

125 executed, 0 reused rule/judge records | 0 manual accepted (unscored; no score-cache credit) | $32.08 rule/judge cost | reconcile exit: 0

Profile: pr / mode full-fallback; selected behaviors: 113, judges: 24. Deferred to scheduled/release coverage: 143 behaviors and 0 changed prompt files. Deferred checks did not run and receive no PR-pass credit.

Full gate restored by:

• test/size-ledger/ship/gsta147-review-depth.json (real unknown dependency): register test/size-ledger/ship/gsta147-review-depth.json under the cases that consume it in test/helpers/touchfiles-data.ts, add it to FREE_ONLY_PR_FILES in scripts/test-pr-profile.ts if no paid case reads it, or add its directory to DERIVABLE_PREFIXES so scripts/pr-dependencies.ts derives its consumers

Derived dependents (scripts/pr-dependencies.ts):

• bin/gstack-doctor-components.sh: 255 case(s), 24 judge(s)
• bin/gstack-review-plan: 255 case(s), 24 judge(s)
• bin/gstack-ship-receipt: 255 case(s), 24 judge(s)
• lib/headless-artifacts.ts: 255 case(s), 24 judge(s)
• lib/review-collect.ts: 255 case(s), 24 judge(s)
• lib/review-depth.ts: 255 case(s), 24 judge(s)
• lib/ship-queue-train.ts: 255 case(s), 24 judge(s)
• lib/ship-queue.ts: 255 case(s), 24 judge(s)
• lib/ship-receipt-depth.ts: 255 case(s), 24 judge(s)
• lib/ship-receipt.ts: 255 case(s), 24 judge(s)

Rule and judge shards

Shard	Automated result	Manual/unscored	Executed	Reused	Status	Cost
llm-judge/skill-llm-eval	24/24	0	24	0	✅	$1.44
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-current	1/1	0	1	0	✅	$0.81
e2e/skill-e2e-coverage-audit	2/2	0	2	0	✅	$0.28
e2e-deploy/skill-e2e-deploy--setup-deploy-workflow	1/1	0	1	0	✅	$0.13
e2e-qa-functional-fix/skill-e2e-qa-functional-fix	2/2	0	2	0	✅	$0.95
e2e/skill-e2e-plan--plan-review-report	1/1	0	1	0	✅	$1.03
e2e-qa-callers/skill-e2e-qa-callers--ship-exploratory-unavailable	1/1	0	1	0	✅	$0.71
e2e-retro/skill-e2e-retro	1/1	0	1	0	✅	$0.33
e2e/skill-e2e-shared-libs-paths--shared-libs-review-prior-coverage	1/1	0	1	0	✅	$0.95
e2e-qa-workflow/skill-e2e-qa-workflow--qa-quick	1/1	0	1	0	✅	$0.48
e2e-review-attribution/skill-e2e-review-attribution	3/3	0	3	0	✅	$0.27
e2e/skill-e2e-ship-skip	1/1	0	1	0	✅	$0.23
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-stale-before	1/1	0	1	0	✅	$0.39
e2e/skill-e2e-shared-libs--shared-libs-review-lifecycle	1/1	0	1	0	✅	$0.89
e2e-review-army/skill-e2e-review-army--review-army-migration-safety	1/1	0	1	0	✅	$0.27
e2e-deploy/skill-e2e-deploy--land-and-deploy-first-run	1/1	0	1	0	✅	$0.32
e2e/skill-e2e-shared-libs--shared-libs-read-only	1/1	0	1	0	✅	$0.24
e2e-deploy/skill-e2e-deploy--land-and-deploy-review-gate	1/1	0	1	0	✅	$0.28
e2e-docsync-spawned/skill-e2e-docsync-spawned	1/1	0	1	0	✅	$0.41
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-stale-after	1/1	0	1	0	✅	$0.41
e2e-safety-codex-consult/skill-e2e-safety-codex-consult	1/1	0	1	0	✅	$0.27
e2e-deploy/skill-e2e-deploy--land-and-deploy-workflow	1/1	0	1	0	✅	$0.56
e2e/skill-e2e-workflow	4/4	0	4	0	✅	$0.96
e2e-review-army/skill-e2e-review-army--review-army-perf-n-plus-one	1/1	0	1	0	✅	$0.2
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-late-result	1/1	0	1	0	✅	$0.34
e2e/skill-e2e-shared-libs-paths--shared-libs-review-index-flags	1/1	0	1	0	✅	$1.01
e2e-deploy/skill-e2e-deploy--benchmark-workflow	1/1	0	1	0	✅	$0.28
e2e/skill-e2e-investigate-owned-completion	1/1	0	1	0	✅	$0.07
e2e-qa-callers/skill-e2e-qa-callers--ship-exploratory-small-cli	0/1	0	1	0	❌	$0.85
e2e-review-army/skill-e2e-review-army--review-army-json-findings	1/1	0	1	0	✅	$0.06
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-store	1/1	0	1	0	✅	$0.79
e2e-triage/skill-e2e-triage	1/1	0	1	0	✅	$0.1
e2e-review-army/skill-e2e-review-army--review-army-delivery-audit	1/1	0	1	0	✅	$0.13
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-timeout-unsettled	1/1	0	1	0	✅	$0.18
e2e-design/skill-e2e-design--design-review-detector-shim	1/1	0	1	0	✅	$0.12
e2e-session-intelligence/skill-e2e-session-intelligence	4/4	0	4	0	✅	$0.25
e2e-safety-pair-agent/skill-e2e-safety-pair-agent	1/1	0	1	0	✅	$0.1
e2e-plan-tune/skill-e2e-plan-tune	1/1	0	1	0	✅	$0.13
e2e-qa-workflow/skill-e2e-qa-workflow--qa-bootstrap	1/1	0	1	0	✅	$0.08
e2e-deploy/skill-e2e-deploy--canary-workflow	1/1	0	1	0	✅	$0.11
e2e/skill-e2e-investigate-owned-termination	2/2	0	2	0	✅	$0.13
e2e/skill-e2e-ship-hook-refresh	1/1	0	1	0	✅	$0.09
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-launch-failure	1/1	0	1	0	✅	$0.16
e2e-design/skill-e2e-design--plan-design-review-no-ui-scope	1/1	0	1	0	✅	$0.09
e2e-third-party-actions/skill-e2e-third-party-actions	5/5	0	5	0	✅	$0.4
e2e/skill-e2e-ship-hook-consent	2/2	0	2	0	✅	$0.2
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-recovery	1/1	0	1	0	✅	$0.37
e2e/skill-e2e-shared-libs--shared-libs-review-revalidation	1/1	0	1	0	✅	$1.85
e2e-review/skill-e2e-review	2/2	0	2	0	✅	$0.32
e2e-cso/skill-e2e-cso	1/1	0	1	0	✅	$0.42
e2e-qa-functional/skill-e2e-qa-functional	2/2	0	2	0	✅	$0.76
e2e-qa-callers/skill-e2e-qa-callers--ship-exploratory-late-input	0/1	0	1	0	❌	$0.93
e2e-learnings/skill-e2e-learnings	1/1	0	1	0	✅	$0.08
e2e-failure-routing/skill-e2e-failure-routing	4/4	0	4	0	✅	$0.37
e2e-qa-workflow/skill-e2e-qa-workflow--qa-only-no-fix	1/1	0	1	0	✅	$0.53
e2e-qa-callers/skill-e2e-qa-callers--review-exploratory-small-cli	1/1	0	1	0	✅	$0.77
e2e-qa-callers/skill-e2e-qa-callers--ship-exploratory-plan-checks	1/1	0	1	0	✅	$0.71
e2e/skill-e2e-test-value	3/3	0	3	0	✅	$0.66
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-failure	1/1	0	1	0	✅	$0.55
e2e-skillify/skill-e2e-skillify	3/3	0	3	0	✅	$0.79
e2e/skill-e2e-shared-libs-paths--shared-libs-review-path-eligibility	1/1	0	1	0	✅	$1.49
e2e-hermetic/skill-e2e-hermetic-canary	2/2	0	2	0	✅	$0.03
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-completion	0/1	0	1	0	❌	$1.45
e2e-design/skill-e2e-design--design-review-plugin-handoff	1/1	0	1	0	✅	$0.12
e2e-review-army/skill-e2e-review-army--review-army-quality-score	1/1	0	1	0	✅	$0.11
e2e-design/skill-e2e-design--design-review-detector-shim-dom	1/1	0	1	0	✅	$0.21
e2e-safety-ship-evidence/skill-e2e-safety-ship-evidence	1/1	0	1	0	✅	$0.12
e2e-browse/skill-e2e-bws	6/6	0	6	0	✅	$0.29
e2e/skill-e2e-shared-libs--shared-libs-unsupported-git	1/1	0	1	0	✅	$0.19
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-missing-asset	1/1	0	1	0	✅	$0.32
e2e-ship-docsync/skill-e2e-ship-docsync--ship-docsync-missing-marker	1/1	0	1	0	✅	$0.16

Fail-closed reconciliation

nc.test.ts#ship-docsync-timeout-unsettled
  slice 16  passed             33s  test/skill-e2e-ask-user-question-format-compliance.test.ts
  slice 16  passed             22s  test/skill-e2e-ship-docsync.test.ts#ship-docsync-launch-failure
  slice 16  passed             26s  test/skill-e2e-session-intelligence.test.ts
  slice 16  passed             32s  test/skill-e2e-ship-hook-consent.test.ts
  slice 16  passed             25s  test/skill-e2e-investigate-owned-termination.test.ts
  slice 16  passed             25s  test/skill-e2e-design.test.ts#design-review-detector-shim
  slice 16  passed             22s  test/skill-e2e-deploy.test.ts#canary-workflow
  slice 16  passed             29s  test/skill-e2e-ship-hook-refresh.test.ts
  slice 16  passed             17s  test/skill-e2e-design.test.ts#plan-design-review-no-ui-scope
  slice 16  passed             26s  test/skill-e2e-review-army.test.ts#review-army-delivery-audit
  slice 16  passed             22s  test/skill-e2e-safety-pair-agent.test.ts
  slice 16  passed             20s  test/skill-e2e-qa-workflow.test.ts#qa-bootstrap
  slice 17  passed             14s  test/skill-e2e-learnings.test.ts
  slice 17  passed             15s  test/skill-e2e-diagram.test.ts
[test:paid] collector final outcomes: 125 executed, 0 reused; 122 passed, 3 failed, 0 manual accepted (unscored; no score-cache credit) (125 attempt records from 71 collectors; every record counts)
[test:paid] deselected testcases, zero credit (316):
[test:paid]    302  sibling case of a case shard (its own shard runs it)
[test:paid]      8  periodic-tier case (this lane runs gate)
[test:paid]      6  Bun hook placeholder of a skipped describe block (not a test)
[test:paid] VERDICT GREEN — lane gate/pr, attempt 1
  rule 112/112 · behavior 1/1 · judge 24/24 · quarantined 0 (0 failing the lane)
  SKIPPED 0 · INFRA 0 · INCOMPLETE 0 · unattributed 0 · ACTION REQUIRED 0
  wall 6m02s · cost $32.74 known + 32 of 139 trial(s) cost unknown
  FLAKY 3 (failed once, passed on its one rule-16 rerun) · 3 rerun execution(s), cost $2.75
[test:paid] flaky (rule 16: failed once, passed on its one rerun; green with this annotation):
  ⚑ ship-docsync-completion  rule  FLAKY: failed in attempt 1 (paid-slice-3-a1), passed on its one rerun in attempt 1 rerun (slice 1 attempt 1); rerun cost $1.12; first failure: assertion: completion fields
  ⚑ ship-exploratory-late-input  rule  FLAKY: failed in attempt 1 (paid-slice-9-a1), passed on its one rerun in attempt 1 rerun (slice 1 attempt 1); rerun cost $0.91; first failure: assertion: expect(received).toEqual(expected)
  ⚑ ship-exploratory-small-cli  rule  FLAKY: failed in attempt 1 (paid-slice-14-a1), passed on its one rerun in attempt 1 rerun (slice 1 attempt 1); rerun cost $0.72; first failure: assertion: expect(received).toEqual(expected)
[test:paid] exercised-path map (shadow, non-blocking): 6 MAP_GAP line(s) over 106 executed case(s) with recorded commands
  MAP_GAP document-release bin/gstack-banned-terms: non-blocking for this PR; the next census refresh widens the map (docs/TESTING_INTERNALS.md#exercised-path-map)
  MAP_GAP plan-review-report bin/gstack-plan-reality: non-blocking for this PR; the next census refresh widens the map (docs/TESTING_INTERNALS.md#exercised-path-map)
  MAP_GAP plan-review-report browse/dist/browse: non-blocking for this PR; the next census refresh widens the map (docs/TESTING_INTERNALS.md#exercised-path-map)
  MAP_GAP plan-review-report design/dist/design: non-blocking for this PR; the next census refresh widens the map (docs/TESTING_INTERNALS.md#exercised-path-map)
  MAP_GAP test-audit-report-only design/dist/design: non-blocking for this PR; the next census refresh widens the map (docs/TESTING_INTERNALS.md#exercised-path-map)
  MAP_GAP timeline-event-flow bin/gstack-skill-end: non-blocking for this PR; the next census refresh widens the map (docs/TESTING_INTERNALS.md#exercised-path-map)
[test:paid] report: every planned shard accounted and passed

---

Sliced lane: planner → duration-packed executors → fail-closed report. Behavior cases run a pre-registered 3-trial panel (PASS at 2/3 with no contract violation); a PASS 2/3 is shown with its failed trial, never as a clean pass. Reused results retain their original provenance and expiry.

Flaky: failed once, passed on its one rerun (rule 16)

⚑ ship-docsync-completion  rule  FLAKY: failed in attempt 1 (paid-slice-3-a1), passed on its one rerun in attempt 1 rerun (slice 1 attempt 1); rerun cost $1.12; first failure: assertion: completion fields
⚑ ship-exploratory-late-input  rule  FLAKY: failed in attempt 1 (paid-slice-9-a1), passed on its one rerun in attempt 1 rerun (slice 1 attempt 1); rerun cost $0.91; first failure: assertion: expect(received).toEqual(expected)
⚑ ship-exploratory-small-cli  rule  FLAKY: failed in attempt 1 (paid-slice-14-a1), passed on its one rerun in attempt 1 rerun (slice 1 attempt 1); rerun cost $0.72; first failure: assertion: expect(received).toEqual(expected)

Each failed and passed on its rerun in run 40805543956 at the attempts named above. Logged for the daily flaky-fix pass in this run's flake-records-a1 artifact (docs/TESTING_INTERNALS.md#flaky-reruns).

Dispatched narrow screen (gstack-ship-queue ready --screen) of merge commit 7ac79f14420a4de7081abe2c045751b0212c9ae3.

[trunk-io]

😎 This pull request was merged.