Merge main e85005482 (v1.93.27.0, train 94) into wave/review-53cfb5bfe011 (rule 19 grant)

Conflicts resolved keeping both sides: bin/gstack-review-dispatch (main's targetedRead record plus the wave's exclude digest on the rendered event); bin/gstack-ship-receipt (main's receiptTargetedAtHead and targetedClassesSince plus the wave's receiptReviewAtHead and receiptExclude; both help paragraphs); docs/troubleshooting.md (main's targeted-refusal fix plus the wave's review_exclude, REVIEW_DIFF_TOO_LARGE, REVIEW_DIFF_UNREADABLE and REVIEW_EXCLUDE_REFERENCE entries); ship/DRIVER.md.tmpl (main's line, DRIVER.md 12,241 B); test/review-packets-removed.test.ts (both replay lists, the wave's placeholder brief-body hashes under main's GST-986 test). test/review-targeted.test.ts follows main's DRIVER wording ([--review-targeted @<f>]), which main's own copy of that check in test/review-targeted-record.test.ts pins.
This commit is contained in:
garrytan committed 2026-11-02 23:50:02 +00:00
commit feeec2e760
74 files changed
+1618 -316

No files matched your search

+115
View File
@@ -0,0 +1,115 @@
{
"calls": [
{
"preReview": true,
"observedAtMs": 76003,
"call": {
"sessionId": "4767bbfe-f301-4d0c-9cfb-eeff94eb0850",
"toolUseId": "toolu_01S2K6zpLEy8wZXVMiG7teSA",
"questions": [
{
"question": "D1 — Review focus for the dashboard plan\nProject/branch/task: main, design review of the User Dashboard Page plan (review-input.md).\nELI10: I've rated this plan 4/10 on design completeness. The data and safety rules are solid, but it never says what the user sees first, how the three panels sit on phone vs desktop, or what the empty/error/toast/modal moments actually say and do. I'll generate visual mockups next, then review all 7 design dimensions. You can narrow that to specific areas instead.\nStakes if we pick wrong: Narrowing too early leaves gaps like toast accessibility or mobile layout for the implementer to guess at, and they ship as generic defaults.\nRecommendation: A because the gaps span hierarchy, states, responsive and a11y, so a partial pass would miss real issues.\nCompleteness: A=10/10, B=6/10\nPros / cons:\nA) Full review, all 7 passes (recommended)\n ✅ Covers hierarchy, states, journey, slop risk, design system, responsive and a11y\n ✅ Every gap gets its own decision so nothing is silently assumed\n ❌ More questions to answer, one per real gap found across the passes\nB) Focus on specific areas only\n ✅ Fewer questions if you only care about, say, states and mobile layout\n ✅ Faster path to an amended plan for the areas you name\n ❌ Unreviewed dimensions stay at their current low rating and ship as guesses\nNet: Full coverage with more decisions, versus speed with known blind spots.",
"header": "Focus",
"multiSelect": false,
"options": [
{
"label": "Full review, all 7 (recommended)",
"description": "Mockups, then all 7 passes. ✅ Covers every dimension. ✅ Each gap decided individually. ❌ More questions overall."
},
{
"label": "Focus on specific areas",
"description": "Name the areas in a note. ✅ Fewer questions. ✅ Faster. ❌ Other dimensions remain unreviewed guesses."
}
]
}
],
"answered": true,
"failed": false,
"answers": {
"D1 — Review focus for the dashboard plan\nProject/branch/task: main, design review of the User Dashboard Page plan (review-input.md).\nELI10: I've rated this plan 4/10 on design completeness. The data and safety rules are solid, but it never says what the user sees first, how the three panels sit on phone vs desktop, or what the empty/error/toast/modal moments actually say and do. I'll generate visual mockups next, then review all 7 design dimensions. You can narrow that to specific areas instead.\nStakes if we pick wrong: Narrowing too early leaves gaps like toast accessibility or mobile layout for the implementer to guess at, and they ship as generic defaults.\nRecommendation: A because the gaps span hierarchy, states, responsive and a11y, so a partial pass would miss real issues.\nCompleteness: A=10/10, B=6/10\nPros / cons:\nA) Full review, all 7 passes (recommended)\n ✅ Covers hierarchy, states, journey, slop risk, design system, responsive and a11y\n ✅ Every gap gets its own decision so nothing is silently assumed\n ❌ More questions to answer, one per real gap found across the passes\nB) Focus on specific areas only\n ✅ Fewer questions if you only care about, say, states and mobile layout\n ✅ Faster path to an amended plan for the areas you name\n ❌ Unreviewed dimensions stay at their current low rating and ship as guesses\nNet: Full coverage with more decisions, versus speed with known blind spots.": "Full review, all 7 (recommended)"
},
"unansweredQuestionIndices": [],
"answeredAt": "2026-11-02T08:39:03.879Z"
}
},
{
"preReview": true,
"observedAtMs": 106367,
"call": {
"sessionId": "4767bbfe-f301-4d0c-9cfb-eeff94eb0850",
"toolUseId": "toolu_01AqE7XDKvovVJAvKF31kq8z",
"questions": [
{
"question": "Want outside design voices before the detailed review?\nD2 — Outside design voices\nProject/branch/task: main, design review of the User Dashboard Page plan.\nELI10: Optionally, two other reviewers read the plan before my 7 passes: Codex checks hard design rules and a litmus scorecard, and an independent Claude subagent checks completeness. Their disagreements get raised in the relevant pass.\nStakes if we pick wrong: Skipping means one reviewer's eye; running costs a few minutes and may only get one voice if Codex is missing.\nRecommendation: No, proceed without — this is a taste call, no strong preference either way\nNote: options differ in kind, not coverage — no completeness score.\nPros / cons: see option descriptions.\nNet: A cross-model check versus the shortest path to a fixed plan.",
"header": "Voices",
"multiSelect": false,
"options": [
{
"label": "Yes, run outside design voices",
"description": "✅ Codex hard-rule + litmus scorecard plus an independent Claude completeness read. ✅ Disagreements surface with both perspectives in the relevant pass. ❌ Adds a few minutes; Codex may be missing here, in which case only the Claude voice runs and it's tagged single-model."
},
{
"label": "No, proceed without (recommended)",
"description": "✅ Straight into the 7 passes with the approved mockup as anchor. ✅ Fewest steps to a fixed plan. ❌ Findings rest on one reviewer; no cross-model check on slop risk or hierarchy."
}
]
}
],
"answered": true,
"failed": false,
"answers": {
"Want outside design voices before the detailed review?\nD2 — Outside design voices\nProject/branch/task: main, design review of the User Dashboard Page plan.\nELI10: Optionally, two other reviewers read the plan before my 7 passes: Codex checks hard design rules and a litmus scorecard, and an independent Claude subagent checks completeness. Their disagreements get raised in the relevant pass.\nStakes if we pick wrong: Skipping means one reviewer's eye; running costs a few minutes and may only get one voice if Codex is missing.\nRecommendation: No, proceed without — this is a taste call, no strong preference either way\nNote: options differ in kind, not coverage — no completeness score.\nPros / cons: see option descriptions.\nNet: A cross-model check versus the shortest path to a fixed plan.": "No, proceed without (recommended)"
},
"unansweredQuestionIndices": [],
"answeredAt": "2026-11-02T08:39:34.676Z"
}
},
{
"preReview": false,
"observedAtMs": 132955,
"call": {
"sessionId": "4767bbfe-f301-4d0c-9cfb-eeff94eb0850",
"toolUseId": "toolu_01PDy4qdecNenP7T1QhxRkFR",
"questions": [
{
"question": "D3 — 1: What does a just-logged-in user see first, second, third?\nProject/branch/task: main, User Dashboard plan, Pass 1 Information Architecture.\nELI10: The plan has three panels but no order. If we don't pick, the engineer puts them in a grid of three equal boxes and nobody knows where to look. Only notifications ask the user to do something, so they're the natural anchor.\nStakes if we pick wrong: Users skim past the one thing needing attention, or the page feels like a generic widget grid.\nRecommendation: 1A because unread notifications are the only actionable content, and Krug says each screen needs one obvious next step.\nNote: options differ in kind, not coverage — no completeness score.\nPros / cons:\nA) Action row, then Notifications primary, Activity secondary (recommended)\n ✅ The one actionable thing (unread items) gets the wide main column and first scan\n ✅ Quick actions sit in the page header as a compact row, one primary button\n ❌ Activity feed is visually demoted; heavy activity users scroll or look right\nB) Activity primary, Notifications in a side rail\n ✅ Suits products where 'what changed' matters more than inbox triage\n ✅ Notifications rail stays visible while scanning activity history\n ❌ Unread items compete in a narrow rail and are easier to ignore\nNet: Anchor on what needs action (notifications) or on what changed (activity).",
"header": "Hierarchy",
"multiSelect": false,
"options": [
{
"label": "1A Notifications primary (recommended)",
"description": "Header + quick-action row first, Notifications wide main column second, Activity secondary column third.",
"preview": "lg (≥1024px)\n+------------------------------------------------+\n| Dashboard [Open workspace] [Review notifications (3)] |\n+-------------------------------+----------------+\n| Notifications 3 unread | Recent activity|\n| [Mark all as read] | • edited ... |\n| ● New comment on Q4 plan 2m | • shared ... |\n| ● ... | • ... |\n| ... | |\n+-------------------------------+----------------+\n 1st: page title + primary action\n 2nd: unread notifications\n 3rd: recent activity"
},
{
"label": "1B Activity primary",
"description": "Header + quick-action row first, Activity wide main column second, Notifications side rail third.",
"preview": "lg (≥1024px)\n+------------------------------------------------+\n| Dashboard [Open workspace] [Review notifications (3)] |\n+-------------------------------+----------------+\n| Recent activity | Notifications |\n| • You edited Q4 roadmap 2m | 3 unread |\n| • ... | [Mark all read]|\n| • ... | ● ... |\n+-------------------------------+----------------+\n 1st: page title + primary action\n 2nd: recent activity\n 3rd: notifications rail"
}
]
}
],
"answered": true,
"failed": false,
"answers": {
"D3 — 1: What does a just-logged-in user see first, second, third?\nProject/branch/task: main, User Dashboard plan, Pass 1 Information Architecture.\nELI10: The plan has three panels but no order. If we don't pick, the engineer puts them in a grid of three equal boxes and nobody knows where to look. Only notifications ask the user to do something, so they're the natural anchor.\nStakes if we pick wrong: Users skim past the one thing needing attention, or the page feels like a generic widget grid.\nRecommendation: 1A because unread notifications are the only actionable content, and Krug says each screen needs one obvious next step.\nNote: options differ in kind, not coverage — no completeness score.\nPros / cons:\nA) Action row, then Notifications primary, Activity secondary (recommended)\n ✅ The one actionable thing (unread items) gets the wide main column and first scan\n ✅ Quick actions sit in the page header as a compact row, one primary button\n ❌ Activity feed is visually demoted; heavy activity users scroll or look right\nB) Activity primary, Notifications in a side rail\n ✅ Suits products where 'what changed' matters more than inbox triage\n ✅ Notifications rail stays visible while scanning activity history\n ❌ Unread items compete in a narrow rail and are easier to ignore\nNet: Anchor on what needs action (notifications) or on what changed (activity).": "1A Notifications primary (recommended)"
},
"unansweredQuestionIndices": [],
"answeredAt": "2026-11-02T08:40:01.258Z"
}
}
],
"provenance": {
"kind": "public native AskUserQuestion calls and answers only (observation.json fingerprints[].nativeCall, byte-exact)",
"runId": 41414111284,
"artifact": "paid-slice-6-a1",
"prNumber": null,
"headSha": "4a280637792b46576474f3349a1ba2e977c5f5c9",
"testedSha": "4a280637792b46576474f3349a1ba2e977c5f5c9",
"recordedOutcome": "ceiling_reached",
"recordedReviewCount": 1,
"recordedStep0Count": 2,
"recordedPassed": false,
"note": "Step 0D titled \"Review focus\" with header Focus, the 4/10 design-completeness rating in its ELI10 body, and an all-7 option that does not lead with the count (\"Full review, all 7 (recommended)\")"
}
}
+5 -13
View File
@@ -521,7 +521,8 @@ chooses it in item 2 and ALREADY_BUMPED derives it in item 1.
runs items 1 and 3–5; each decision left to you is a `STOP:` naming its rerun flag
(`--bump`, `--accept-incomplete`, `--version <v>`, `--keep-current`, `--answer`).
`ship: queue_mode stamp-at-merge` replaces items 1–4.
`VERSION: skipped` or `VERSION: <v> kept (…; no write)` ends Step 12 with
`NEW_VERSION` unset: Step 13 follows its `NEXT:`; Steps 18 and 20 take NO_VERSION's branch.
1. **Classify state** — pure reader, never writes:
```bash
@@ -642,19 +643,10 @@ Make bisectable commits; if already committed, continue to Step 16. Never create
migrations may stand alone. Under 50 lines across fewer than 4 files may be one commit.
2. Order dependencies first: infrastructure → models/services → controllers/views.
Each commit works alone (no broken imports or missing code).
Group VERSION + CHANGELOG + TODOS.md after the feature commits.
3. Use `<type>: <summary>` (feat/fix/chore/refactor/docs) and a brief body; fixing a
user-named issue adds a `Fixes: #<n>` trailer. Only the final
VERSION/CHANGELOG commit gets the release version and co-author trailer. No Git tag:
```bash
git commit -m "$(cat <<'EOF'
chore: bump version and changelog (vX.Y.Z.W)
Co-Authored-By: Claude <noreply@anthropic.com>
EOF
)"
```
user-named issue adds a `Fixes: #<n>` trailer. Leave VERSION, CHANGELOG/fragment, TODOS.md
and the digest uncommitted: Step 16's `finish` commits them last with
`--trailer 'Co-Authored-By: Claude <noreply@anthropic.com>'`. Do not add a Git tag.
---
+5 -13
View File
@@ -520,7 +520,8 @@ chooses it in item 2 and ALREADY_BUMPED derives it in item 1.
runs items 1 and 3–5; each decision left to you is a `STOP:` naming its rerun flag
(`--bump`, `--accept-incomplete`, `--version <v>`, `--keep-current`, `--answer`).
`ship: queue_mode stamp-at-merge` replaces items 1–4.
`VERSION: skipped` or `VERSION: <v> kept (…; no write)` ends Step 12 with
`NEW_VERSION` unset: Step 13 follows its `NEXT:`; Steps 18 and 20 take NO_VERSION's branch.
1. **Classify state** — pure reader, never writes:
```bash
@@ -641,19 +642,10 @@ Make bisectable commits; if already committed, continue to Step 16. Never create
migrations may stand alone. Under 50 lines across fewer than 4 files may be one commit.
2. Order dependencies first: infrastructure → models/services → controllers/views.
Each commit works alone (no broken imports or missing code).
Group VERSION + CHANGELOG + TODOS.md after the feature commits.
3. Use `<type>: <summary>` (feat/fix/chore/refactor/docs) and a brief body; fixing a
user-named issue adds a `Fixes: #<n>` trailer. Only the final
VERSION/CHANGELOG commit gets the release version and co-author trailer. No Git tag:
```bash
git commit -m "$(cat <<'EOF'
chore: bump version and changelog (vX.Y.Z.W)
Co-Authored-By: Claude <noreply@anthropic.com>
EOF
)"
```
user-named issue adds a `Fixes: #<n>` trailer. Leave VERSION, CHANGELOG/fragment, TODOS.md
and the digest uncommitted: Step 16's `finish` commits them last with
`--trailer 'Co-Authored-By: Claude <noreply@anthropic.com>'`. Do not add a Git tag.
---
+5 -13
View File
@@ -520,7 +520,8 @@ chooses it in item 2 and ALREADY_BUMPED derives it in item 1.
runs items 1 and 3–5; each decision left to you is a `STOP:` naming its rerun flag
(`--bump`, `--accept-incomplete`, `--version <v>`, `--keep-current`, `--answer`).
`ship: queue_mode stamp-at-merge` replaces items 1–4.
`VERSION: skipped` or `VERSION: <v> kept (…; no write)` ends Step 12 with
`NEW_VERSION` unset: Step 13 follows its `NEXT:`; Steps 18 and 20 take NO_VERSION's branch.
1. **Classify state** — pure reader, never writes:
```bash
@@ -644,19 +645,10 @@ Make bisectable commits; if already committed, continue to Step 16. Never create
migrations may stand alone. Under 50 lines across fewer than 4 files may be one commit.
2. Order dependencies first: infrastructure → models/services → controllers/views.
Each commit works alone (no broken imports or missing code).
Group VERSION + CHANGELOG + TODOS.md after the feature commits.
3. Use `<type>: <summary>` (feat/fix/chore/refactor/docs) and a brief body; fixing a
user-named issue adds a `Fixes: #<n>` trailer. Only the final
VERSION/CHANGELOG commit gets the release version and co-author trailer. No Git tag:
```bash
git commit -m "$(cat <<'EOF'
chore: bump version and changelog (vX.Y.Z.W)
Co-Authored-By: OpenAI Codex <noreply@openai.com>
EOF
)"
```
user-named issue adds a `Fixes: #<n>` trailer. Leave VERSION, CHANGELOG/fragment, TODOS.md
and the digest uncommitted: Step 16's `finish` commits them last with
`--trailer 'Co-Authored-By: OpenAI Codex <noreply@openai.com>'`. Do not add a Git tag.
---
+5 -13
View File
@@ -522,7 +522,8 @@ chooses it in item 2 and ALREADY_BUMPED derives it in item 1.
runs items 1 and 3–5; each decision left to you is a `STOP:` naming its rerun flag
(`--bump`, `--accept-incomplete`, `--version <v>`, `--keep-current`, `--answer`).
`ship: queue_mode stamp-at-merge` replaces items 1–4.
`VERSION: skipped` or `VERSION: <v> kept (…; no write)` ends Step 12 with
`NEW_VERSION` unset: Step 13 follows its `NEXT:`; Steps 18 and 20 take NO_VERSION's branch.
1. **Classify state** — pure reader, never writes:
```bash
@@ -646,19 +647,10 @@ Make bisectable commits; if already committed, continue to Step 16. Never create
migrations may stand alone. Under 50 lines across fewer than 4 files may be one commit.
2. Order dependencies first: infrastructure → models/services → controllers/views.
Each commit works alone (no broken imports or missing code).
Group VERSION + CHANGELOG + TODOS.md after the feature commits.
3. Use `<type>: <summary>` (feat/fix/chore/refactor/docs) and a brief body; fixing a
user-named issue adds a `Fixes: #<n>` trailer. Only the final
VERSION/CHANGELOG commit gets the release version and co-author trailer. No Git tag:
```bash
git commit -m "$(cat <<'EOF'
chore: bump version and changelog (vX.Y.Z.W)
Co-Authored-By: Factory Droid <droid@users.noreply.github.com>
EOF
)"
```
user-named issue adds a `Fixes: #<n>` trailer. Leave VERSION, CHANGELOG/fragment, TODOS.md
and the digest uncommitted: Step 16's `finish` commits them last with
`--trailer 'Co-Authored-By: Factory Droid <droid@users.noreply.github.com>'`. Do not add a Git tag.
---
+6
View File
@@ -0,0 +1,6 @@
# garrytan/gstack-evals#141: native adversarial findings #4 and #6, head a752186 (tree 0ac70e872724875710ccc47f9c93b7b30a58604a)
Copied verbatim from /home/user/.capy/work/ship-141/run/p2/out/adversarial-native.txt
4. **driver.ts:352, 561 and 618 (the C1 ruling): skill-eval sessions now end with background Agents still running and leave no structured record.** run-host.ts:352 passes no wait. Claude Code 2.1.314 runs Agent calls in the background by default. So a skill-eval `--host claude` session ends and is graded on its "agents launched, waiting" turn, which is the original GST-734 bug. Its record also has no `session_end`, only a line in cli-stderr.log. The CREW rule asks for a per-arm count of sessions that ended with tasks pending, and these records can't give that count. Grades for skills that hand work to Agents may be deflated with nothing marking them. Old cached sessions do not mix in, because adapterSha256 hashes driver.ts. The fix: keep emitting `session_end` (cause `tasks_pending`, `waited_ms` 0) when no wait is set, and leave the hold opt-in. That reverses the C1 test's `"session_end" in r.outcome === false` assertion, so the C1 owner needs to agree. FIXABLE. Severity: Medium. Blocks: no
6. **driver.ts:404-407: the timeout branch for an idle session sets `overdue = true` without checking `stop`.** Here is the case. A background agent's refused request sets a stop while the main agent is idle, such as `cap_stop`, or an infra failure does. The timer then fires during abort teardown. The record ends up with a `cap_stop` or infra termination but `session_end.cause` "deadline". The `watch` interval already checks `stop`. Fix: `else if (idleSince !== null && !stop)`. FIXABLE. Severity: Low. Blocks: no
+1 -1
View File
@@ -1 +1 @@
{"note":"parseFindings labels at main 82e9d7ec (GST-976): every input the review-dispatch and prose-voice-floor tests pass it, every reader fixture here, and generated Severity field shapes. Regenerate only from a frozen main.","main":"82e9d7ec7a2c222f20b04aaf5b965fa50d7b6bdb","entries":[{"source":"pinned","reader":"adversarial-native","output":"FIXABLE: race in a.ts:3\nSeverity: Medium\nBlocks: no","main":["CRITICAL"]},{"source":"pinned","reader":"adversarial-native","output":"INVESTIGATE: odd lock in b.ts:9\nSeverity: Low\nBlocks: yes","main":["CRITICAL"]},{"source":"pinned","reader":"adversarial-native","output":"INVESTIGATE: odd lock in b.ts:9\nBlocks: no","main":["INVESTIGATE"]},{"source":"pinned","reader":"adversarial-native","output":"FIXABLE: race in a.ts:3. Severity: Medium. Blocks: no","main":["CRITICAL"]},{"source":"pinned","reader":"adversarial-native","output":"1. FIXABLE: race in a.ts:3 — Severity: Low, Blocks: no","main":["CRITICAL"]},{"source":"pinned","reader":"adversarial-native","output":"- FIXABLE: race in a.ts:3\n- Severity: Medium\n- Blocks: no","main":["CRITICAL"]},{"source":"pinned","reader":"outside-voice","output":"1. Drops the last row (q.ts:7). Severity: Critical. Blocks: no","main":["Critical"]},{"source":"pinned","reader":"outside-voice","output":"1. Cache evicts sessions (c.ts:4)\n Severity: High","main":["High"]},{"source":"pinned","reader":"outside-voice","output":"1. Retry storm (r.ts:2)\n Severity: Medium\n Blocks: yes","main":["CRITICAL"]},{"source":"pinned","reader":"outside-voice","output":"1. Index drift (i.ts:5). Severity: Low. Blocks: Yes, it loses the row","main":["CRITICAL"]},{"source":"pinned","reader":"outside-voice","output":"1. Naming (n.ts:1)\n Severity: Low\n Blocks: no","main":["Low"]},{"source":"pinned","reader":"specialist/testing","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\nNO FINDINGS\n","main":[]},{"source":"pinned","reader":"specialist/testing","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\n{\"severity\":\"CRITICAL\",\"path\":\"a.ts\",\"line\":3,\"category\":\"logic\",\"summary\":\"off by one\",\"specialist\":\"testing\"}\n{\"severity\":\"INFORMATIONAL\",\"path\":\"a.ts\",\"line\":9,\"category\":\"style\",\"summary\":\"name\",\"specialist\":\"testing\"}\n","main":["CRITICAL","INFORMATIONAL"]},{"source":"pinned","reader":"outside-voice","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\nFindings:\n1. **High**: retry loop never backs off (q.ts:78)\n2. Low - log line is noisy\nRecommendation: Fix the retry because it DoSes the pool\n","main":["High","Low"]},{"source":"pinned","reader":"specialist/testing","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\n{\"severity\":\"CRITICAL\",\"path\":\"a.ts\",\"line\":3,\"category\":\"logic\",\"summary\":\"x\",\"specialist\":\"testing\"}\n","main":["CRITICAL"]},{"source":"pinned","reader":"red-team","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\n{\"severity\":\"CRITICAL\",\"path\":\"b.ts\",\"line\":1,\"category\":\"x\",\"summary\":\"y\",\"specialist\":\"red-team\"}\n","main":["CRITICAL"]},{"source":"pinned","reader":"specialist/testing","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\n{\"severity\":\"CRITICAL\",\"path\":\"a.ts\",\"line\":3,\"category\":\"logic\",\"summary\":\"x\",\"specialist\":\"testing\"}\n{\"severity\":\"INFORMATIONAL\",\"path\":\"b.ts\",\"line\":1,\"category\":\"n\",\"summary\":\"y\",\"specialist\":\"testing\"}\n","main":["CRITICAL","INFORMATIONAL"]},{"source":"pinned","reader":"r","output":"NO FINDINGS\n","main":[]},{"source":"pinned","reader":"r","output":"## High\n1. The auth check skips webhooks (a.ts:4)\n - Impact: nested evidence\n **Fix:**\n- second one\n## Notes\n- not a finding\n**Critical findings**\n- real one\n**Recommendations**\n- consider renaming foo\n","main":["High","High","Critical"]},{"source":"pinned","reader":"r","output":"1. **Unbounded retry in queue.ts:78** (FIXABLE): loops on 429\n2. `auth.ts:42` skips token expiry check — INVESTIGATE\n### CRITICAL: missing auth check in api.ts:10\n#### [P1] token not expired\n### High: authorization bypass (auth.ts:14)\n- **HIGH**: leak\n| High | race in queue.ts:4 |\nClassification: INVESTIGATE\n7. **Informational: reads every file**","main":["FIXABLE","INVESTIGATE","CRITICAL","P1","High","HIGH","High","INVESTIGATE","Informational"]},{"source":"pinned","reader":"r","output":"FIXABLE: race in a.ts:3\nINVESTIGATE: odd lock\nRecommendation: Fix the race because x\n","main":["FIXABLE","INVESTIGATE"]},{"source":"pinned","reader":"outside-voice","output":"No CRITICAL issues except the token leak in log.ts:3","main":["CRITICAL"]},{"source":"pinned","reader":"fix-check","output":"No CRITICAL issues except the token leak in lLine truncated
{"note":"parseFindings labels at main 82e9d7ec (GST-976): every input the review-dispatch and prose-voice-floor tests pass it, every reader fixture here, and generated Severity field shapes. Regenerate only from a frozen main.","main":"82e9d7ec7a2c222f20b04aaf5b965fa50d7b6bdb","entries":[{"source":"pinned","reader":"adversarial-native","output":"FIXABLE: race in a.ts:3\nSeverity: Medium\nBlocks: no","main":["CRITICAL"]},{"source":"pinned","reader":"adversarial-native","output":"INVESTIGATE: odd lock in b.ts:9\nSeverity: Low\nBlocks: yes","main":["CRITICAL"]},{"source":"pinned","reader":"adversarial-native","output":"INVESTIGATE: odd lock in b.ts:9\nBlocks: no","main":["INVESTIGATE"]},{"source":"pinned","reader":"adversarial-native","output":"FIXABLE: race in a.ts:3. Severity: Medium. Blocks: no","main":["CRITICAL"]},{"source":"pinned","reader":"adversarial-native","output":"1. FIXABLE: race in a.ts:3 — Severity: Low, Blocks: no","main":["CRITICAL"]},{"source":"pinned","reader":"adversarial-native","output":"- FIXABLE: race in a.ts:3\n- Severity: Medium\n- Blocks: no","main":["CRITICAL"]},{"source":"pinned","reader":"outside-voice","output":"1. Drops the last row (q.ts:7). Severity: Critical. Blocks: no","main":["Critical"]},{"source":"pinned","reader":"outside-voice","output":"1. Cache evicts sessions (c.ts:4)\n Severity: High","main":["High"]},{"source":"pinned","reader":"outside-voice","output":"1. Retry storm (r.ts:2)\n Severity: Medium\n Blocks: yes","main":["CRITICAL"]},{"source":"pinned","reader":"outside-voice","output":"1. Index drift (i.ts:5). Severity: Low. Blocks: Yes, it loses the row","main":["CRITICAL"]},{"source":"pinned","reader":"outside-voice","output":"1. Naming (n.ts:1)\n Severity: Low\n Blocks: no","main":["Low"]},{"source":"pinned","reader":"specialist/testing","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\nNO FINDINGS\n","main":[]},{"source":"pinned","reader":"specialist/testing","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\n{\"severity\":\"CRITICAL\",\"path\":\"a.ts\",\"line\":3,\"category\":\"logic\",\"summary\":\"off by one\",\"specialist\":\"testing\"}\n{\"severity\":\"INFORMATIONAL\",\"path\":\"a.ts\",\"line\":9,\"category\":\"style\",\"summary\":\"name\",\"specialist\":\"testing\"}\n","main":["CRITICAL","INFORMATIONAL"]},{"source":"pinned","reader":"outside-voice","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\nFindings:\n1. **High**: retry loop never backs off (q.ts:78)\n2. Low - log line is noisy\nRecommendation: Fix the retry because it DoSes the pool\n","main":["High","Low"]},{"source":"pinned","reader":"specialist/testing","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\n{\"severity\":\"CRITICAL\",\"path\":\"a.ts\",\"line\":3,\"category\":\"logic\",\"summary\":\"x\",\"specialist\":\"testing\"}\n","main":["CRITICAL"]},{"source":"pinned","reader":"red-team","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\n{\"severity\":\"CRITICAL\",\"path\":\"b.ts\",\"line\":1,\"category\":\"x\",\"summary\":\"y\",\"specialist\":\"red-team\"}\n","main":["CRITICAL"]},{"source":"pinned","reader":"specialist/testing","output":"REVIEWED_TREE: d0dd42d6a88fbc2a6abca4f6b2dfde7e35520c5a /home/user/.capy/work/gst976/wt\n{\"severity\":\"CRITICAL\",\"path\":\"a.ts\",\"line\":3,\"category\":\"logic\",\"summary\":\"x\",\"specialist\":\"testing\"}\n{\"severity\":\"INFORMATIONAL\",\"path\":\"b.ts\",\"line\":1,\"category\":\"n\",\"summary\":\"y\",\"specialist\":\"testing\"}\n","main":["CRITICAL","INFORMATIONAL"]},{"source":"pinned","reader":"r","output":"NO FINDINGS\n","main":[]},{"source":"pinned","reader":"r","output":"## High\n1. The auth check skips webhooks (a.ts:4)\n - Impact: nested evidence\n **Fix:**\n- second one\n## Notes\n- not a finding\n**Critical findings**\n- real one\n**Recommendations**\n- consider renaming foo\n","main":["High","High","Critical"]},{"source":"pinned","reader":"r","output":"1. **Unbounded retry in queue.ts:78** (FIXABLE): loops on 429\n2. `auth.ts:42` skips token expiry check — INVESTIGATE\n### CRITICAL: missing auth check in api.ts:10\n#### [P1] token not expired\n### High: authorization bypass (auth.ts:14)\n- **HIGH**: leak\n| High | race in queue.ts:4 |\nClassification: INVESTIGATE\n7. **Informational: reads every file**","main":["FIXABLE","INVESTIGATE","CRITICAL","P1","High","HIGH","High","INVESTIGATE","Informational"]},{"source":"pinned","reader":"r","output":"FIXABLE: race in a.ts:3\nINVESTIGATE: odd lock\nRecommendation: Fix the race because x\n","main":["FIXABLE","INVESTIGATE"]},{"source":"pinned","reader":"outside-voice","output":"No CRITICAL issues except the token leak in log.ts:3","main":["CRITICAL"]},{"source":"pinned","reader":"fix-check","output":"No CRITICAL issues except the token leak in lLine truncated