Conversation
…g + transition matrix tests (順位 85+86 / Bundle g-1) §A-2 Phase 5 dogfood P-1 (Bundle g-1)。3 PR 連続観測 (PR #119/#120/#121) で 発覚した monitor の誤 approved 判定を fix。 順位 85 (Tier 1, T1-1): - compute_verdict() に review_state guard 追加。CodeRabbit が未投稿 (review_state: not_found) もしくは進行中 (pending) のときは findings の有無に関わらず判定保留。空 findings を 'no problems' と誤同一視する false negative を防止。 順位 86 (Tier 2, T2-4): - mod tests に (action, review_state, findings) → verdict transition matrix を 12 unit test で網羅: - parked_rate_limit / parked_review_recheck (action 優先) - not_found + 空 / pending + 空 / not_found + findings (順位 85 fix) - success + 空 / minor / critical / high / major - skipped (skip_coderabbit 経路) - coderabbit field None (初期 state) - 既存 should_resume_wakeup_* テスト 7 件と並存、no regression build + deploy 済 (.claude/cli-pr-monitor.exe を release build で再生成)。
|
ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughPR監視ロジックのcompute_verdict関数がCodeRabbitのreview_stateフィールドを考慮するように拡張されました。review_stateが「not_found」または「pending」の場合、新しい「pending」判定を返します。既存のparked actionロジックは優先度を維持し、テストスイートが判定の優先順位ルールと新しい動作を検証します。 ChangesReview State Verdict Logic
Estimated code review effort🎯 2 (Simple) | ⏱️ ~12 minutes Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
7 tasks
aloekun
added a commit
that referenced
this pull request
May 7, 2026
…P-1/P-2 ledger (順位 7 / §A-2 P-3) (#127) * feat(hooks-post-tool-linter): PowerShell (?i) flag validation + §A-2 P-1/P-2 ledger 記録 (順位 7 / §A-2 P-3) §A-2 Phase 5 dogfood P-3 (順位 7)。PR #91 で発覚した PowerShell case-insensitive regex の構造的落とし穴 (no-empty-powershell-catch / no-silent-error-action で (?i) 欠落 → CR Major 指摘) を機械強制で再発防止。 順位 7 (Tier 1, T1-1): - find_powershell_rules_missing_case_insensitive_flag() 関数追加: extensions に ps1 を含む rule で pattern に (?i) が無いものを ID リストで返す - load_custom_rules() で起動時 warn (本番運用層) - 7 unit tests (異種 violator / valid / mixed-ext / case-insensitive ext / multi-violator / non-ps1 ignore) + 1 deployed-config 全 rule 検証 test (CI 検出層) - 既存 no-ephemeral-todo-reference rule の pattern に (?i) を追加 (deployed validation を pass、Windows file path 大文字小文字混在対応の副次効果あり) - ~/.claude/rules/common/code-review.md § Custom lint rule patterns 追加 (PR #91 a15b263 fix の経緯 + 順位 7 機械強制の事後参照) §A-2 計測ログ (P-1/P-2) 記録: - P-1 (PR #125): findings: 0 (CR APPROVE no comments)、classifier 未起動 → dogfood 不発 - P-2 (PR #126): findings: 1 (CR Nitpick、review body <details> block 内、 check-ci-coderabbit 抽出漏れ)、手動 synthetic finding で classifier 実行 → action=human_review / action_confidence=0.0 / fallback=length_contract、 agreement: 1/1 (100%)、latency: 6.4s/件 (>5s 目標)、fallback: 1/1 - 既知 system gap: check-ci-coderabbit が review body の <details> Nitpick を 抽出しない (post-pr-monitor が classifier に渡す入力経路に欠落) build + deploy 済 (.claude/hooks-post-tool-linter.exe)。 全 82 tests pass、regression なし。 * docs(todo): 順位 7 (PowerShell (?i) flag 自動検証) 完了に伴い削除
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
compute_verdict()に review_state guard 追加 — CodeRabbit がnot_found/pendingのとき判定保留(action, review_state, findings) → verdictの transition matrix を 12 unit test で網羅Changes
src/cli-pr-monitor/src/stages/monitor.rscompute_verdict()に early-return guard を追加:parked_*の match 分岐後、severity ベース判定の前。coderabbitが None (初期 state / skip_coderabbit) のときは guard 通過、既存挙動を保持mod tests末尾に 12 unit test 追加:should_resume_wakeup_*7 件と並存、regression なし.claude/cli-pr-monitor.exepnpm build:cli-pr-monitorで release build を再生成し配備バグ再現条件
not_found[]pending[]not_found[major]success[]skipped[]§A-2 dogfood (P-1) 観測ポイント
本 PR の post-pr-monitor で
[classifier](mistral:7b) が CodeRabbit findings を分類する初の dogfood サイクル。計測対象:action一致率 — 目標 ≥80%human_reviewfallback の比率 — 目標 ≤20%normalized_issue言語制約違反率: 英語 8 文字以上連続検出 — 目標 ≤10%PR merge 後に
.takt/dogfood-pr-NNN-classified.jsonで findings + classified を保存し、docs/local-llm-offload-analysis.md§A-2 計測ログに追記する。Test plan
cargo test -p cli-pr-monitor verdict --quiet— 12/12 passcargo test -p cli-pr-monitor --quiet— 163 active passed (7 ignored は事前)、regression なし.claude/cli-pr-monitor.exe)関連
docs/todo5.md詳細エントリdocs/local-llm-offload-analysis.md(本 PR が dogfood の初回起動)Summary by CodeRabbit
リリースノート
改善
テスト