← スキル一覧に戻る

debug-deep
by dthompson-jti
⭐ 0🍴 0📅 2026年1月15日
SKILL.md
name: debug-deep description: Tree-of-thoughts reasoning for elusive, intermittent bugs. Use when standard debugging fails or the bug is hard to reproduce.
Debug Deep
Tree-of-thoughts reasoning for elusive bugs.
When to Use
- Intermittent or hard-to-reproduce bugs
- Standard debugging failed
- Race conditions suspected
- Multiple potential causes
Artifacts
DEBUG-PLAN-[BUG-ID].mdtask.mdupdated viatask_boundary
Approach
Phase 1: False Premise Challenge
List every assumption in the bug report. Test each as True/False before proceeding:
- Assumption 1: [e.g., "Bug occurs on every load"]
- Assumption 2: [e.g., "Data is correctly formatted"]
Phase 2: Tree of Thoughts (Multi-Hypothesis)
Generate 4-6 hypotheses with dual investigation paths:
| # | Hypothesis | Confidence | Path A | Path B | Selected |
|---|---|---|---|---|---|
| 1 | ... | High | Check X with grep | Instrument Y | A (faster) |
| 2 | ... | Medium | Trace via debugger | Reproduce isolated | B (reliable) |
Phase 3: Investigation & Fix Plan
- Primary Fix Plan: Main approach
- Reflexion (Security Engineer): List 3 ways fix might fail, introduce vulnerabilities, or degrade performance
- Fallback Plan: If primary fails
Phase 4: Systemic Analysis
- Does this bug represent a class of problems?
- Should a new rule be added to prevent recurrence?
Constraints
- Non-Linear Exploration: Don't converge prematurely
- Must have fallback plan
- If fix fails, iterate (max 3 attempts)
スコア
総合スコア
50/100
リポジトリの品質指標に基づく評価
✓SKILL.md
SKILL.mdファイルが含まれている
+20
○LICENSE
ライセンスが設定されている
0/10
○説明文
100文字以上の説明がある
0/10
○人気
GitHub Stars 100以上
0/15
○最近の活動
3ヶ月以内に更新がある
0/10
○フォーク
10回以上フォークされている
0/5
✓Issue管理
オープンIssueが50未満
+5
✓言語
プログラミング言語が設定されている
+5
○タグ
1つ以上のタグが設定されている
0/5
レビュー
💬
レビュー機能は近日公開予定です