mandrel 2.111.0 → 2.112.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (178) hide show
  1. package/.agents/agents/acceptance-critic.md +14 -0
  2. package/.agents/agents/auditor.md +14 -0
  3. package/.agents/agents/plan-critic.md +5 -0
  4. package/.agents/agents/story-worker.md +4 -1
  5. package/.agents/docs/agentrc-reference.json +22 -10
  6. package/.agents/docs/configuration.md +351 -37
  7. package/.agents/docs/execution-reference.md +11 -0
  8. package/.agents/rules/git-conventions-reference.md +7 -0
  9. package/.agents/schemas/agentrc.schema.json +174 -14
  10. package/.agents/schemas/baselines/baseline-envelope.schema.json +16 -4
  11. package/.agents/schemas/baselines/bundle-size.schema.json +1 -0
  12. package/.agents/schemas/baselines/coverage.schema.json +1 -0
  13. package/.agents/schemas/baselines/crap.schema.json +37 -10
  14. package/.agents/schemas/baselines/duplication.schema.json +1 -0
  15. package/.agents/schemas/baselines/maintainability.schema.json +1 -0
  16. package/.agents/schemas/baselines/mutation.schema.json +1 -0
  17. package/.agents/schemas/crap-report.schema.json +6 -5
  18. package/.agents/schemas/story-deliver-terminal.schema.json +41 -2
  19. package/.agents/schemas/story-progress.schema.json +1 -1
  20. package/.agents/schemas/validation-evidence.schema.json +2 -2
  21. package/.agents/scripts/boot-sweep.js +41 -3
  22. package/.agents/scripts/coverage-capture.js +37 -3
  23. package/.agents/scripts/coverage-ratchet.js +106 -0
  24. package/.agents/scripts/evidence-gate.js +84 -21
  25. package/.agents/scripts/lib/audit-baselines/read.js +8 -1
  26. package/.agents/scripts/lib/audit-baselines/trend.js +2 -1
  27. package/.agents/scripts/lib/baseline-loader.js +4 -2
  28. package/.agents/scripts/lib/baselines/codec.js +57 -0
  29. package/.agents/scripts/lib/baselines/crap-preview-incremental.js +8 -3
  30. package/.agents/scripts/lib/baselines/crap-preview-scan.js +10 -16
  31. package/.agents/scripts/lib/baselines/crap-updater-cli.js +4 -13
  32. package/.agents/scripts/lib/baselines/kernel.js +19 -1
  33. package/.agents/scripts/lib/baselines/kinds/_crap-codec.js +171 -0
  34. package/.agents/scripts/lib/baselines/kinds/_crap-read.js +56 -16
  35. package/.agents/scripts/lib/baselines/kinds/_shared-metric.js +0 -59
  36. package/.agents/scripts/lib/baselines/kinds/crap.js +283 -319
  37. package/.agents/scripts/lib/baselines/preview-gates.js +29 -10
  38. package/.agents/scripts/lib/baselines/reader.js +36 -9
  39. package/.agents/scripts/lib/baselines/refresh-service.js +79 -18
  40. package/.agents/scripts/lib/baselines/seat-missing.js +173 -7
  41. package/.agents/scripts/lib/baselines/writer.js +36 -21
  42. package/.agents/scripts/lib/changed-files.js +70 -6
  43. package/.agents/scripts/lib/child-exec.js +10 -0
  44. package/.agents/scripts/lib/claude/agents-projection.js +5 -5
  45. package/.agents/scripts/lib/close-lock-events.js +8 -3
  46. package/.agents/scripts/lib/close-validation/ci-deferral.js +40 -0
  47. package/.agents/scripts/lib/close-validation/gates.js +111 -35
  48. package/.agents/scripts/lib/close-validation/process.js +16 -7
  49. package/.agents/scripts/lib/close-validation/projections/crap.js +43 -10
  50. package/.agents/scripts/lib/close-validation/runner.js +55 -23
  51. package/.agents/scripts/lib/close-validation/skip-branch-test.js +144 -0
  52. package/.agents/scripts/lib/close-validation/test-delta.js +33 -8
  53. package/.agents/scripts/lib/codex/agents-projection.js +6 -0
  54. package/.agents/scripts/lib/config/agentrc-overlay.js +9 -5
  55. package/.agents/scripts/lib/config/explain.js +5 -2
  56. package/.agents/scripts/lib/config/gates/coverage.schema.js +45 -1
  57. package/.agents/scripts/lib/config/gates/crap-incremental-coverage.schema.js +1 -1
  58. package/.agents/scripts/lib/config/git-roots.js +38 -4
  59. package/.agents/scripts/lib/config/handoff.js +24 -0
  60. package/.agents/scripts/lib/config/quality.js +26 -13
  61. package/.agents/scripts/lib/config/worktree-isolation.js +2 -5
  62. package/.agents/scripts/lib/config-resolver.js +22 -2
  63. package/.agents/scripts/lib/config-settings-schema-delivery.js +87 -4
  64. package/.agents/scripts/lib/config-settings-schema-quality.js +36 -1
  65. package/.agents/scripts/lib/config-settings-schema.js +1 -1
  66. package/.agents/scripts/lib/coverage-capture-affected.js +1 -0
  67. package/.agents/scripts/lib/coverage-capture-delta.js +164 -32
  68. package/.agents/scripts/lib/coverage-capture-own-fix.js +232 -0
  69. package/.agents/scripts/lib/coverage-capture-usage.js +2 -2
  70. package/.agents/scripts/lib/coverage-capture.js +15 -0
  71. package/.agents/scripts/lib/coverage-deferral.js +96 -0
  72. package/.agents/scripts/lib/coverage-ratchet.js +187 -0
  73. package/.agents/scripts/lib/crap-baseline-join.js +5 -5
  74. package/.agents/scripts/lib/crap-engine.js +3 -3
  75. package/.agents/scripts/lib/crap-method-identity.js +104 -48
  76. package/.agents/scripts/lib/escomplex-anchor.js +135 -0
  77. package/.agents/scripts/lib/escomplex-kernel.js +30 -1
  78. package/.agents/scripts/lib/failure-rows.js +121 -0
  79. package/.agents/scripts/lib/full-suite-lock.js +44 -4
  80. package/.agents/scripts/lib/full-suite-queue.js +89 -13
  81. package/.agents/scripts/lib/generated/agentrc-validator.js +8 -2
  82. package/.agents/scripts/lib/gh-env.js +3 -6
  83. package/.agents/scripts/lib/gh-pr-rest.js +1 -0
  84. package/.agents/scripts/lib/git/base-sync-probe.js +182 -0
  85. package/.agents/scripts/lib/git-branch-lifecycle.js +9 -0
  86. package/.agents/scripts/lib/observability/close-telemetry.js +37 -2
  87. package/.agents/scripts/lib/observability/handoff-telemetry.js +170 -0
  88. package/.agents/scripts/lib/observability/review-dispositions.js +23 -6
  89. package/.agents/scripts/lib/observability/runtime-friction.js +17 -4
  90. package/.agents/scripts/lib/observability/source-classifier.js +3 -0
  91. package/.agents/scripts/lib/observability/story-outcome.js +393 -0
  92. package/.agents/scripts/lib/orchestration/check-baselines/phases/compare.js +2 -1
  93. package/.agents/scripts/lib/orchestration/check-baselines/phases/evaluate.js +19 -1
  94. package/.agents/scripts/lib/orchestration/check-baselines/phases/fail-detail.js +84 -0
  95. package/.agents/scripts/lib/orchestration/check-baselines/phases/pipeline.js +14 -5
  96. package/.agents/scripts/lib/orchestration/check-baselines/phases/refresh-ack.js +3 -2
  97. package/.agents/scripts/lib/orchestration/check-baselines/phases/report.js +26 -6
  98. package/.agents/scripts/lib/orchestration/code-review.js +223 -19
  99. package/.agents/scripts/lib/orchestration/handoff-gates.js +129 -62
  100. package/.agents/scripts/lib/orchestration/handoff-rescore.js +169 -0
  101. package/.agents/scripts/lib/orchestration/handoff-suite-of-record.js +45 -0
  102. package/.agents/scripts/lib/orchestration/preflight-lanes.js +208 -0
  103. package/.agents/scripts/lib/orchestration/review-base-ref.js +40 -17
  104. package/.agents/scripts/lib/orchestration/review-deposit.js +38 -12
  105. package/.agents/scripts/lib/orchestration/review-depth.js +21 -0
  106. package/.agents/scripts/lib/orchestration/review-history.js +124 -20
  107. package/.agents/scripts/lib/orchestration/review-providers/claude-verdict.js +23 -8
  108. package/.agents/scripts/lib/orchestration/review-providers/code-review.js +82 -39
  109. package/.agents/scripts/lib/orchestration/review-providers/critical-verifier.js +64 -11
  110. package/.agents/scripts/lib/orchestration/review-providers/degraded-gates.js +98 -3
  111. package/.agents/scripts/lib/orchestration/review-providers/finding-identity.js +6 -2
  112. package/.agents/scripts/lib/orchestration/review-providers/findings-renderer.js +55 -8
  113. package/.agents/scripts/lib/orchestration/review-providers/review-guidance.js +1 -0
  114. package/.agents/scripts/lib/orchestration/review-providers/review-provider-factory.js +42 -15
  115. package/.agents/scripts/lib/orchestration/review-providers/types.js +6 -0
  116. package/.agents/scripts/lib/orchestration/single-story-close/failed-terminal.js +10 -1
  117. package/.agents/scripts/lib/orchestration/single-story-close/gate-log.js +10 -0
  118. package/.agents/scripts/lib/orchestration/single-story-close/phases/base-sync.js +7 -4
  119. package/.agents/scripts/lib/orchestration/single-story-close/phases/close-validation.js +15 -3
  120. package/.agents/scripts/lib/orchestration/single-story-close/phases/code-review.js +59 -2
  121. package/.agents/scripts/lib/orchestration/single-story-close/phases/confirm-merge.js +6 -0
  122. package/.agents/scripts/lib/orchestration/single-story-close/phases/finish.js +15 -4
  123. package/.agents/scripts/lib/orchestration/single-story-close/phases/land-review-record.js +213 -0
  124. package/.agents/scripts/lib/orchestration/single-story-close/phases/normalize-pr-title.js +9 -12
  125. package/.agents/scripts/lib/orchestration/single-story-close/phases/open-pr.js +26 -12
  126. package/.agents/scripts/lib/orchestration/single-story-close/phases/post-land.js +51 -26
  127. package/.agents/scripts/lib/orchestration/single-story-close/phases/pre-push.js +3 -2
  128. package/.agents/scripts/lib/orchestration/single-story-close/phases/pull-request.js +50 -17
  129. package/.agents/scripts/lib/orchestration/single-story-close/phases/push.js +59 -11
  130. package/.agents/scripts/lib/orchestration/single-story-close/phases/reconcile-done.js +1 -0
  131. package/.agents/scripts/lib/orchestration/single-story-close/phases/review-block.js +31 -13
  132. package/.agents/scripts/lib/orchestration/single-story-close/phases/review-halt.js +29 -6
  133. package/.agents/scripts/lib/orchestration/single-story-close/phases/review-outcome.js +10 -12
  134. package/.agents/scripts/lib/orchestration/single-story-close/phases/review-override.js +19 -4
  135. package/.agents/scripts/lib/orchestration/single-story-close/result.js +30 -13
  136. package/.agents/scripts/lib/orchestration/single-story-close/review-overlap.js +73 -39
  137. package/.agents/scripts/lib/orchestration/single-story-close/runner.js +11 -3
  138. package/.agents/scripts/lib/orchestration/story-close/baseline-upward-writeback.js +3 -2
  139. package/.agents/scripts/lib/orchestration/story-close/duplication-close-rescore.js +20 -8
  140. package/.agents/scripts/lib/orchestration/story-deliver-terminal.js +7 -0
  141. package/.agents/scripts/lib/orchestration/story-handoff.js +774 -196
  142. package/.agents/scripts/lib/orchestration/ticketing/reads.js +2 -0
  143. package/.agents/scripts/lib/pr-body-lines.js +32 -0
  144. package/.agents/scripts/lib/single-story/ci-ratchet-gate.js +140 -0
  145. package/.agents/scripts/lib/single-story/confirm-and-land.js +17 -7
  146. package/.agents/scripts/lib/single-story/confirm-merge.js +12 -3
  147. package/.agents/scripts/lib/single-story-sweep/report-stalled-executing.js +202 -0
  148. package/.agents/scripts/lib/standalone-ratchets.js +37 -0
  149. package/.agents/scripts/lib/suite-admission.js +20 -8
  150. package/.agents/scripts/lib/suite-sizing.js +146 -0
  151. package/.agents/scripts/lib/workspace-provisioner.js +2 -2
  152. package/.agents/scripts/merge-baseline.js +14 -3
  153. package/.agents/scripts/providers/github/comments.js +19 -3
  154. package/.agents/scripts/providers/github.js +1 -0
  155. package/.agents/scripts/review-dispositions-report.js +1 -1
  156. package/.agents/scripts/role-scope-guard.js +267 -0
  157. package/.agents/scripts/single-story-close.js +4 -4
  158. package/.agents/scripts/single-story-confirm-merge.js +31 -5
  159. package/.agents/scripts/single-story-init.js +24 -4
  160. package/.agents/scripts/story-handoff.js +3 -3
  161. package/.agents/scripts/story-outcomes-report.js +231 -0
  162. package/.agents/scripts/story-review-compute.js +9 -5
  163. package/.agents/scripts/update-coverage-baseline.js +12 -2
  164. package/.agents/scripts/update-crap-baseline.js +9 -1
  165. package/.agents/workflows/helpers/code-review.md +63 -9
  166. package/.agents/workflows/helpers/deliver-digest.md +53 -54
  167. package/.agents/workflows/helpers/deliver-reference.md +233 -67
  168. package/.agents/workflows/helpers/deliver-story-reference.md +19 -8
  169. package/.agents/workflows/helpers/deliver-story.md +7 -2
  170. package/.agents/workflows/helpers/parallel-tooling.md +29 -0
  171. package/docs/CHANGELOG.md +46 -0
  172. package/lib/cli/doctor-checks/coverage-ratchet.js +54 -0
  173. package/lib/cli/doctor-checks/effective-agentrc.js +23 -0
  174. package/lib/cli/doctor-checks/llm-review.js +2 -11
  175. package/lib/cli/doctor-checks/sync-parity.js +5 -1
  176. package/lib/cli/doctor-checks/test-credit-path.js +7 -2
  177. package/lib/cli/registry.js +5 -0
  178. package/package.json +1 -1
@@ -7,9 +7,16 @@
7
7
  * @typedef {import('./review-providers/types.js').Finding} Finding
8
8
  * @typedef {{ id: string, severity: string, provider: string|null,
9
9
  * class: string|null, instances: { file: string, line?: number }[],
10
- * ungrounded?: string, unverified?: string }} HistoryFinding
11
- * @typedef {{ runs: { headSha: string, findings: HistoryFinding[] }[],
10
+ * ungrounded?: string, unverified?: string, carried?: string,
11
+ * title?: string, body?: string, file?: string, line?: number,
12
+ * claim?: object }} HistoryFinding
13
+ * @typedef {{ id: string, status: 'still-open'|'fixed'|'refuted',
14
+ * reason?: string }} EarlierOutcome
15
+ * @typedef {{ runs: { headSha: string, findings: HistoryFinding[],
16
+ * earlierCriticals?: EarlierOutcome[] }[],
12
17
  * overridden: string[], overrideReason?: string }} ReviewHistory
18
+ * @typedef {{ id: string, finding: HistoryFinding, headSha: string,
19
+ * status: 'open'|'fixed'|'refuted'|'overridden', reason?: string }} EarlierCritical
13
20
  */
14
21
 
15
22
  import nodeFs from 'node:fs';
@@ -28,6 +35,8 @@ const KIND = 'story-review-history';
28
35
  const LINE_WINDOW = 10;
29
36
  /** Same bound as the `review-block-overridden` friction signal. */
30
37
  const OVERRIDE_REASON_LIMIT = 500;
38
+ /** A stored CRITICAL body is clipped to this many characters. */
39
+ const BODY_LIMIT = 4000;
31
40
 
32
41
  /**
33
42
  * @param {number} storyId
@@ -86,6 +95,31 @@ function updateReviewHistory(storyId, update, deps = {}) {
86
95
  );
87
96
  }
88
97
 
98
+ /** Keep the set entries of `fields`; an empty string is unset. */
99
+ const present = (fields) =>
100
+ Object.fromEntries(
101
+ Object.entries(fields).filter(([, v]) => v != null && v !== ''),
102
+ );
103
+
104
+ const flags = (f) =>
105
+ present({
106
+ ungrounded: f.ungrounded,
107
+ unverified: f.unverified,
108
+ carried: f.carried,
109
+ });
110
+
111
+ /** What the verifier needs to re-check a CRITICAL without its original run. */
112
+ const criticalContent = (f) =>
113
+ f.severity === 'critical'
114
+ ? present({
115
+ title: f.title,
116
+ body: typeof f.body === 'string' ? f.body.slice(0, BODY_LIMIT) : null,
117
+ file: f.file,
118
+ line: f.line,
119
+ claim: f.claim,
120
+ })
121
+ : {};
122
+
89
123
  /** @param {Finding} f @returns {HistoryFinding} */
90
124
  function toHistoryFinding(f) {
91
125
  return {
@@ -94,27 +128,89 @@ function toHistoryFinding(f) {
94
128
  provider: f.provider ?? null,
95
129
  class: f.class ?? null,
96
130
  instances: findingInstances(f),
97
- ...(f.ungrounded ? { ungrounded: f.ungrounded } : {}),
98
- ...(f.unverified ? { unverified: f.unverified } : {}),
131
+ ...flags(f),
132
+ ...criticalContent(f),
99
133
  };
100
134
  }
101
135
 
102
136
  /**
103
137
  * @param {number} storyId
104
- * @param {{ headSha: string, findings: Finding[] }} run
138
+ * @param {{ headSha: string, findings: Finding[],
139
+ * earlierCriticals?: EarlierOutcome[] }} run
105
140
  * @param {{ config?: object, fsImpl?: typeof nodeFs }} [deps]
106
141
  */
107
- export function appendReviewRun(storyId, { headSha, findings }, deps) {
108
- updateReviewHistory(
109
- storyId,
110
- (h) => ({
111
- ...h,
112
- runs: [...h.runs, { headSha, findings: findings.map(toHistoryFinding) }],
113
- }),
114
- deps,
142
+ export function appendReviewRun(
143
+ storyId,
144
+ { headSha, findings, earlierCriticals },
145
+ deps,
146
+ ) {
147
+ const run = toHistoryRun(headSha, findings, earlierCriticals);
148
+ updateReviewHistory(storyId, (h) => ({ ...h, runs: [...h.runs, run] }), deps);
149
+ }
150
+
151
+ function toHistoryRun(headSha, findings, earlierCriticals) {
152
+ const run = { headSha, findings: findings.map(toHistoryFinding) };
153
+ if (Array.isArray(earlierCriticals)) run.earlierCriticals = earlierCriticals;
154
+ return run;
155
+ }
156
+
157
+ const list = (v) => (Array.isArray(v) ? v : []);
158
+ const RESOLVED = new Set(['fixed', 'refuted']);
159
+
160
+ /** Raised as `critical` and survived grounding and verification. */
161
+ const survived = (f) =>
162
+ f?.severity === 'critical' && !f.unverified && !f.ungrounded;
163
+
164
+ /**
165
+ * Every CRITICAL that survived grounding and verification in some run, in
166
+ * first-raised order, with its current status. A run's `fixed`/`refuted`
167
+ * outcome closes it; surviving again re-opens it; overridden ids end
168
+ * `overridden`. A carried record keeps the head it was first raised at.
169
+ *
170
+ * @param {ReviewHistory} history
171
+ * @returns {EarlierCritical[]}
172
+ */
173
+ export function earlierCriticals(history) {
174
+ const byId = new Map();
175
+ for (const run of list(history?.runs)) {
176
+ list(run?.earlierCriticals).forEach((o) => {
177
+ resolveEntry(byId, o);
178
+ });
179
+ list(run?.findings)
180
+ .filter(survived)
181
+ .forEach((f) => {
182
+ openEntry(byId, f, run.headSha);
183
+ });
184
+ }
185
+ return markOverridden([...byId.values()], list(history?.overridden));
186
+ }
187
+
188
+ /** Overridden ids end `overridden`. */
189
+ function markOverridden(entries, ids) {
190
+ const overridden = new Set(ids);
191
+ return entries.map((e) =>
192
+ overridden.has(e.id) ? { ...e, status: 'overridden' } : e,
115
193
  );
116
194
  }
117
195
 
196
+ /** A run's `fixed`/`refuted` outcome closes an entry. */
197
+ function resolveEntry(byId, outcome) {
198
+ const entry = byId.get(outcome?.id);
199
+ if (!entry || !RESOLVED.has(outcome.status)) return;
200
+ Object.assign(entry, { status: outcome.status, reason: outcome.reason });
201
+ }
202
+
203
+ /** A surviving CRITICAL (re-)opens its entry; carried keeps the first head. */
204
+ function openEntry(byId, finding, headSha) {
205
+ const prev = byId.get(finding.id);
206
+ byId.set(finding.id, {
207
+ id: finding.id,
208
+ finding,
209
+ headSha: finding.carried && prev ? prev.headSha : headSha,
210
+ status: 'open',
211
+ });
212
+ }
213
+
118
214
  /**
119
215
  * @param {number} storyId
120
216
  * @param {string[]} ids - the overridden CRITICALs.
@@ -148,6 +244,19 @@ function overlaps(now, before) {
148
244
  return now.some((a) => before.some((b) => sameLocation(a, b)));
149
245
  }
150
246
 
247
+ /**
248
+ * Whether `finding` is `earlier` raised again: same id, overlapping location.
249
+ *
250
+ * @param {Finding} finding
251
+ * @param {HistoryFinding} earlier
252
+ */
253
+ export function raisedAgain(finding, earlier) {
254
+ return (
255
+ findingId(finding) === earlier?.id &&
256
+ overlaps(findingInstances(finding), earlier.instances ?? [])
257
+ );
258
+ }
259
+
151
260
  /**
152
261
  * Mark each finding an earlier run raised (same id, overlapping location) as
153
262
  * a `repeat`. A repeat of an overridden CRITICAL is downgraded so it does not
@@ -163,13 +272,8 @@ export function markRepeats(findings, history) {
163
272
  );
164
273
  const overridden = new Set(history.overridden);
165
274
  return findings.map((f) => {
166
- const id = findingId(f);
167
- const instances = findingInstances(f);
168
- const seen = earlier.some(
169
- (e) => e?.id === id && overlaps(instances, e.instances ?? []),
170
- );
171
- if (!seen) return f;
172
- if (f.severity === 'critical' && overridden.has(id)) {
275
+ if (!earlier.some((e) => raisedAgain(f, e))) return f;
276
+ if (f.severity === 'critical' && overridden.has(findingId(f))) {
173
277
  return {
174
278
  ...f,
175
279
  severity: 'suggestion',
@@ -187,11 +187,21 @@ function toFindings(entries) {
187
187
  .filter(Boolean);
188
188
  }
189
189
 
190
+ const resultTextOf = (envelope) =>
191
+ typeof envelope?.result === 'string' ? envelope.result : '';
192
+
193
+ function isUsageLimited(result, envelope) {
194
+ const text = [result.stderr, resultTextOf(envelope), result.stdout];
195
+ return USAGE_LIMIT_RE.test(text.join('\n'));
196
+ }
197
+
198
+ function isRetryable(result, envelope) {
199
+ return Number.isInteger(result.status) && !isUsageLimited(result, envelope);
200
+ }
201
+
190
202
  function failureReason(result, envelope) {
191
- const resultText =
192
- typeof envelope?.result === 'string' ? envelope.result : '';
193
- const text = [result.stderr, resultText, result.stdout].join('\n');
194
- if (USAGE_LIMIT_RE.test(text)) {
203
+ const resultText = resultTextOf(envelope);
204
+ if (isUsageLimited(result, envelope)) {
195
205
  return 'the claude CLI account is out of quota (usage limit reached)';
196
206
  }
197
207
  if (result.status !== 0) {
@@ -237,19 +247,24 @@ function resultTextFindings(text) {
237
247
  /**
238
248
  * Verdict resolution: (a) a successful envelope's `structured_output`,
239
249
  * (b) the fallback parser over its `result` text, (c) no verdict.
250
+ * `retryable` excludes a usage limit and a `status: null` timeout.
240
251
  *
241
- * @param {{ status: number, stdout: string, stderr: string }} result
242
- * @returns {{ findings: Finding[] } | { reason: string }}
252
+ * @param {{ status: number|null, stdout: string, stderr: string }} result
253
+ * @returns {{ findings: Finding[] } | { reason: string, retryable: boolean }}
243
254
  */
244
255
  export function resolveVerdict(result) {
245
256
  const envelope = parseEnvelope(result.stdout);
257
+ const retryable = isRetryable(result, envelope);
246
258
  if (!isSuccess(result, envelope)) {
247
- return { reason: noVerdictReason(result, envelope) };
259
+ return { reason: noVerdictReason(result, envelope), retryable };
248
260
  }
249
261
  const entries =
250
262
  structuredFindings(envelope.structured_output) ??
251
263
  resultTextFindings(envelope.result);
252
264
  return entries
253
265
  ? { findings: toFindings(entries) }
254
- : { reason: 'the reviewer output did not parse as a findings verdict' };
266
+ : {
267
+ reason: 'the reviewer output did not parse as a findings verdict',
268
+ retryable,
269
+ };
255
270
  }
@@ -1,6 +1,6 @@
1
1
  /**
2
- * review-providers/code-review.js — a low-effort model bug review of the
3
- * Story diff through `claude --print --effort low`. It asks only for
2
+ * review-providers/code-review.js — a model bug review of the Story diff
3
+ * through `claude --print --effort <input.effort ?? low>`. It asks only for
4
4
  * merge-blocking problems, so every parsed finding is `critical` and halts
5
5
  * close before auto-merge. The reviewer is handed the diff range, the Story
6
6
  * id and the diff text — never `acceptance[]` or the self-eval verdict: this
@@ -10,14 +10,15 @@
10
10
  * OS temp cwd (no CLAUDE.md discovery), user-only settings, no MCP, no tools.
11
11
  *
12
12
  * A missing CLI throws at construction (the chain entry is optional). A
13
- * review with no verdict records one degraded gate via `getDegradations()`
14
- * and returns no findings, so "never ran" cannot read as "found nothing".
13
+ * transient failure is retried once; a review still with no verdict records
14
+ * one degraded gate via `getDegradations()` and returns no findings.
15
15
  * A diff over the cap is reviewed chunk by chunk (file-aligned, sequential,
16
16
  * findings merged). Source sections pack before tests and docs, and every
17
17
  * chunk is told it is a slice and handed the whole changed-file manifest so
18
- * a file changed in another chunk never reads as unchanged. What went unchecked — a truncated file, files past the
19
- * chunk bound, a chunk with no verdict, a capped `<callers>` block — is
20
- * recorded beside the findings, naming the files.
18
+ * a file changed in another chunk never reads as unchanged. What went
19
+ * unchecked — a truncated file, files past the chunk bound, a chunk with no
20
+ * verdict (`coverage-partial`), a capped `<callers>` block
21
+ * (`context-reduced`) — is recorded beside the findings, naming the files.
21
22
  *
22
23
  * @typedef {import('./types.js').Finding} Finding
23
24
  * @typedef {import('./types.js').ReviewInput} ReviewInput
@@ -28,7 +29,10 @@
28
29
  import fs from 'node:fs';
29
30
  import os from 'node:os';
30
31
  import path from 'node:path';
31
- import { spawnCaptureAsync } from '../../child-exec.js';
32
+ import {
33
+ spawnCaptureAsync,
34
+ spawnCaptureNullableAsync,
35
+ } from '../../child-exec.js';
32
36
  import { gitSpawn } from '../../git-utils.js';
33
37
  import { PROJECT_ROOT } from '../../project-root.js';
34
38
  import { buildCallersSection } from './callers-context.js';
@@ -43,10 +47,11 @@ import { renderDepthDirective } from './review-depth.js';
43
47
  import { renderGuidanceBlock } from './review-guidance.js';
44
48
  import { probeClaudeCli } from './security-review.js';
45
49
 
46
- const CLAUDE_ARGS = Object.freeze([
50
+ /** @param {string} effort */
51
+ const claudeArgs = (effort) => [
47
52
  '--print',
48
53
  '--effort',
49
- 'low',
54
+ effort,
50
55
  '--output-format',
51
56
  'json',
52
57
  '--json-schema',
@@ -56,7 +61,7 @@ const CLAUDE_ARGS = Object.freeze([
56
61
  '--strict-mcp-config',
57
62
  '--tools',
58
63
  '',
59
- ]);
64
+ ];
60
65
  const AUTH_ARGS = Object.freeze(['auth', 'status', '--json']);
61
66
  const INVOKE_TIMEOUT_MS = 10 * 60 * 1000;
62
67
  const AUTH_TIMEOUT_MS = 15 * 1000;
@@ -239,19 +244,22 @@ function quoteWindowsArg(arg) {
239
244
  * @param {{ cwd: string, input?: string, timeout: number, run?: Function, platform: string }} opts
240
245
  * @returns {Promise<{ status: number, stdout: string, stderr: string }>}
241
246
  */
242
- export function runClaude(args, { cwd, input, timeout, run, platform }) {
247
+ export function runClaude(args, opts) {
248
+ return spawnCaptureAsync('claude', ...claudeSpawn(args, opts));
249
+ }
250
+
251
+ /** `runClaude`, but a timeout, kill or spawn error stays `status: null`. */
252
+ function runClaudeNullable(args, opts) {
253
+ return spawnCaptureNullableAsync('claude', ...claudeSpawn(args, opts));
254
+ }
255
+
256
+ /** The argv and spawn options both runners share. */
257
+ function claudeSpawn(args, { cwd, input, timeout, run, platform }) {
243
258
  const shell = platform === 'win32';
244
- return spawnCaptureAsync(
245
- 'claude',
259
+ return [
246
260
  shell ? args.map(quoteWindowsArg) : [...args],
247
- {
248
- cwd,
249
- input: input ?? '',
250
- shell,
251
- timeout,
252
- ...(run ? { run } : {}),
253
- },
254
- );
261
+ { cwd, input: input ?? '', shell, timeout, ...(run ? { run } : {}) },
262
+ ];
255
263
  }
256
264
 
257
265
  /**
@@ -280,8 +288,8 @@ function runAuthStatus(runDir, deps) {
280
288
  });
281
289
  }
282
290
 
283
- function runReviewer(runDir, prompt, deps) {
284
- return runClaude(CLAUDE_ARGS, {
291
+ function runReviewer(runDir, prompt, effort, deps) {
292
+ return runClaudeNullable(claudeArgs(effort), {
285
293
  cwd: runDir,
286
294
  input: prompt,
287
295
  timeout: INVOKE_TIMEOUT_MS,
@@ -293,22 +301,35 @@ function runReviewer(runDir, prompt, deps) {
293
301
  /**
294
302
  * Preflight, then the review, both from a fresh OS temp dir removed after.
295
303
  *
296
- * @returns {Promise<{ findings: Finding[] } | { reason: string }>}
304
+ * @returns {Promise<{ findings: Finding[] } | { reason: string, retryable?: boolean }>}
297
305
  */
298
- async function reviewInTempDir(prompt, deps) {
306
+ async function reviewInTempDir(prompt, effort, deps) {
299
307
  const runDir = fs.mkdtempSync(path.join(os.tmpdir(), 'mandrel-code-review-'));
300
308
  try {
301
309
  const auth = readAuthStatus(await runAuthStatus(runDir, deps));
302
310
  if (!auth.ok) return auth;
303
311
  deps.logger?.info?.(
304
- `[code-review] Invoking claude --print --effort low (${auth.detail})...`,
312
+ `[code-review] Invoking claude --print --effort ${effort} (${auth.detail})...`,
305
313
  );
306
- return resolveVerdict(await runReviewer(runDir, prompt, deps));
314
+ return resolveVerdict(await runReviewer(runDir, prompt, effort, deps));
307
315
  } finally {
308
316
  fs.rmSync(runDir, { recursive: true, force: true });
309
317
  }
310
318
  }
311
319
 
320
+ /** One invocation, re-run once when its failure is transient-shaped. */
321
+ async function reviewWithRetry(prompt, effort, deps) {
322
+ const first = await reviewInTempDir(prompt, effort, deps);
323
+ if (!first.retryable) return first;
324
+ deps.logger?.warn?.(
325
+ `[code-review] No verdict (${first.reason}); retrying once.`,
326
+ );
327
+ return reviewInTempDir(prompt, effort, deps);
328
+ }
329
+
330
+ /** @param {ReviewInput} input */
331
+ const effortOf = (input) => input.effort ?? 'low';
332
+
312
333
  function assertCliPresent(probeFn) {
313
334
  if (probeFn()) return;
314
335
  throw new Error(
@@ -339,18 +360,38 @@ export function createCodeReviewProviderForRegistry(deps = {}) {
339
360
  let degradations = [];
340
361
 
341
362
  /** Record a not-checked slice without replacing the findings. */
342
- function noteDegradation(reason) {
343
- degradations.push({ ...DEGRADATION_BASE, reason });
363
+ function noteDegradation(reason, kind) {
364
+ degradations.push({ ...DEGRADATION_BASE, kind, reason });
344
365
  deps.logger?.warn?.(`[code-review] Not checked: ${reason}.`);
345
366
  }
346
367
 
347
368
  function settle(verdict) {
348
369
  if (!('reason' in verdict)) return verdict.findings;
349
- degradations.push({ ...DEGRADATION_BASE, reason: verdict.reason });
370
+ degradations.push({
371
+ ...DEGRADATION_BASE,
372
+ kind: 'verdict-missing',
373
+ reason: verdict.reason,
374
+ });
350
375
  deps.logger?.warn?.(`[code-review] No verdict: ${verdict.reason}.`);
351
376
  return [];
352
377
  }
353
378
 
379
+ /**
380
+ * No chunk with a verdict is a review with none: one record naming every
381
+ * file. Otherwise each chunk without one is a partial gap.
382
+ */
383
+ function noteChunkGaps(failed, total) {
384
+ if (failed.length > 0 && failed.length === total) {
385
+ const reasons = [...new Set(failed.map((f) => f.reason))].join('; ');
386
+ const files = failed.map((f) => f.files).join(', ');
387
+ noteDegradation(`no verdict for ${files}: ${reasons}`, 'verdict-missing');
388
+ return;
389
+ }
390
+ for (const { files, reason } of failed) {
391
+ noteDegradation(`no verdict for ${files}: ${reason}`, 'coverage-partial');
392
+ }
393
+ }
394
+
354
395
  /**
355
396
  * The `<callers>` block, guidance, the slice note and the manifest ride every chunk,
356
397
  * so they come out of the budget; chunk numbers are one digit, so the
@@ -376,15 +417,18 @@ export function createCodeReviewProviderForRegistry(deps = {}) {
376
417
  noteDegradation(
377
418
  `review bounded at ${MAX_REVIEW_CHUNKS} chunks: ${overflow.length} ` +
378
419
  `file(s) not reviewed: ${overflow.join(', ')}`,
420
+ 'coverage-partial',
379
421
  );
380
422
  }
381
423
  const findings = [];
424
+ const failed = [];
382
425
  for (const [index, chunk] of chunks.entries()) {
383
426
  const files = chunk.sections.map((s) => s.file).join(', ');
384
427
  if (chunk.size > budget) {
385
428
  noteDegradation(
386
429
  `${files} truncated at ${budget} characters: ` +
387
430
  `${chunk.size - budget} characters not reviewed`,
431
+ 'coverage-partial',
388
432
  );
389
433
  }
390
434
  const prompt = buildPrompt(
@@ -394,13 +438,11 @@ export function createCodeReviewProviderForRegistry(deps = {}) {
394
438
  budget,
395
439
  sliceNote(index + 1, chunks.length) + manifest,
396
440
  );
397
- const verdict = await reviewInTempDir(prompt, runDeps);
398
- if ('reason' in verdict) {
399
- noteDegradation(`no verdict for ${files}: ${verdict.reason}`);
400
- } else {
401
- findings.push(...verdict.findings);
402
- }
441
+ const verdict = await reviewWithRetry(prompt, effortOf(input), runDeps);
442
+ if ('reason' in verdict) failed.push({ files, reason: verdict.reason });
443
+ else findings.push(...verdict.findings);
403
444
  }
445
+ noteChunkGaps(failed, chunks.length);
404
446
  return dedupeFindings(findings);
405
447
  }
406
448
 
@@ -417,13 +459,14 @@ export function createCodeReviewProviderForRegistry(deps = {}) {
417
459
  cwd: PROJECT_ROOT,
418
460
  gitSpawnFn,
419
461
  });
420
- if (callers.reason) noteDegradation(callers.reason);
462
+ if (callers.reason) noteDegradation(callers.reason, 'context-reduced');
421
463
  if (diff.text.length > MAX_DIFF_CHARS) {
422
464
  return reviewInChunks(input, diff.text, callers.text);
423
465
  }
424
466
  return settle(
425
- await reviewInTempDir(
467
+ await reviewWithRetry(
426
468
  buildPrompt(input, diff.text, callers.text),
469
+ effortOf(input),
427
470
  runDeps,
428
471
  ),
429
472
  );
@@ -256,6 +256,18 @@ function boundDegradation(skipped) {
256
256
  ];
257
257
  }
258
258
 
259
+ function voterContext(args) {
260
+ return {
261
+ baseRef: args.baseRef,
262
+ headRef: args.headRef,
263
+ cwd: args.cwd ?? PROJECT_ROOT,
264
+ spawnFn: args.spawnFn,
265
+ platform: args.platform ?? process.platform,
266
+ gitSpawnFn: args.gitSpawnFn ?? gitSpawn,
267
+ cliPresent: (args.probeFn ?? probeClaudeCli)(),
268
+ };
269
+ }
270
+
259
271
  /**
260
272
  * Verify every `critical` finding; findings below critical are never sent.
261
273
  * One verifier at `light`/`standard`, three at `deep`.
@@ -278,15 +290,7 @@ async function verifyCriticals(args) {
278
290
  }
279
291
  const checked = criticals.slice(0, MAX_VERIFIED_CRITICALS);
280
292
  const voters = args.depth === 'deep' ? 3 : 1;
281
- const ctx = {
282
- baseRef: args.baseRef,
283
- headRef: args.headRef,
284
- cwd: args.cwd ?? PROJECT_ROOT,
285
- spawnFn: args.spawnFn,
286
- platform: args.platform ?? process.platform,
287
- gitSpawnFn: args.gitSpawnFn ?? gitSpawn,
288
- cliPresent: (args.probeFn ?? probeClaudeCli)(),
289
- };
293
+ const ctx = voterContext(args);
290
294
  args.logger?.info?.(
291
295
  `[code-review] Verifying ${checked.length} critical(s) with ${voters} verifier(s) each...`,
292
296
  );
@@ -313,6 +317,56 @@ async function verifyCriticals(args) {
313
317
  };
314
318
  }
315
319
 
320
+ /** Whether the review gets a verifier: built-in, or injected seams. */
321
+ const hasVerifier = (opts) => Boolean(opts.verifier || !opts.reviewProvider);
322
+
323
+ /**
324
+ * Re-check earlier CRITICALs at the current head: one call each whatever the
325
+ * depth, the same pool and cap as `verifyCriticals`. Each result is the voter
326
+ * outcome; past the cap or with no verifier the outcome is `skipped`.
327
+ *
328
+ * @param {{ findings: Finding[], reviewInput: { baseRef: string,
329
+ * headRef: string }, opts: object }} args
330
+ * @returns {Promise<{ results: { outcome: string, reason: string }[],
331
+ * degradations: GateDegradation[], verification: VerificationSummary }>}
332
+ */
333
+ export async function recheckCriticals({ findings, reviewInput, opts }) {
334
+ const verification = emptyVerification();
335
+ if (findings.length === 0 || !hasVerifier(opts)) {
336
+ const reason = 'no verifier for this review';
337
+ return {
338
+ results: findings.map(() => ({ outcome: 'skipped', reason })),
339
+ degradations: [],
340
+ verification,
341
+ };
342
+ }
343
+ const checked = findings.slice(0, MAX_VERIFIED_CRITICALS);
344
+ const ctx = voterContext({
345
+ baseRef: reviewInput.baseRef,
346
+ headRef: reviewInput.headRef,
347
+ cwd: opts.cwd,
348
+ gitSpawnFn: opts.gitSpawnFn,
349
+ ...opts.verifier,
350
+ });
351
+ const votes = await runPooled(
352
+ checked.map((f) => () => runVoter(f, ctx)),
353
+ MAX_IN_FLIGHT,
354
+ );
355
+ for (const v of votes) verification[v.outcome] += 1;
356
+ verification.attempted = votes.length;
357
+ return {
358
+ results: findings.map(
359
+ (_, i) =>
360
+ votes[i] ?? { outcome: 'skipped', reason: 'verification cap reached' },
361
+ ),
362
+ degradations: [
363
+ ...checked.flatMap((f, i) => voteDegradations(f, [votes[i]])),
364
+ ...boundDegradation(findings.slice(MAX_VERIFIED_CRITICALS)),
365
+ ],
366
+ verification,
367
+ };
368
+ }
369
+
316
370
  /**
317
371
  * The pipeline step: ground every finding, then verify the CRITICALs left.
318
372
  * An injected review provider gets no verifier unless `opts.verifier` (the
@@ -337,8 +391,7 @@ export async function vetFindings({
337
391
  degradations,
338
392
  gitSpawnFn: opts.gitSpawnFn,
339
393
  });
340
- const verify =
341
- opts.verifier || !opts.reviewProvider ? verifyCriticals : skipVerification;
394
+ const verify = hasVerifier(opts) ? verifyCriticals : skipVerification;
342
395
  const verified = await verify({
343
396
  findings: grounded.findings,
344
397
  baseRef,