メニュー
Use when encountering a bug, regression, failing command, or unexpected behavior before attempting a fix.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use only when the user explicitly allows bounded unattended execution of an approved Alloy task.
Use before planning when requirements are ambiguous, large, user-visible, or likely to cause rework unless decisions are locked.
Use when implementing an approved Alloy plan. Maintains progress.md, enforces TDD/debug discipline, and stops on blocked gates.
Use when a task needs broad source understanding before planning or execution.
Use when turning context or a clear user request into an implementation-ready plan with acceptance criteria, files, commands, gates, risks, and handoff.
Automated end-to-end QA — read acceptance criteria (Azure DevOps or manual), generate Playwright test plan, execute with screenshots/video/GIF, produce acceptance report. Uses Playwright CLI for deterministic, reportable execution.
| name | alloy-debug |
| description | Use when encountering a bug, regression, failing command, or unexpected behavior before attempting a fix. |
Use this skill before fixing failures. The goal is root cause, not motion.
No fix before root-cause investigation.
Update .alloy/tasks/<task-id>/progress.md:
## Gate
- [x] debug - <timestamp> | root cause documented
## Findings
### Debug - <timestamp>
- Symptom:
- Reproduction:
- Root cause:
- Fix:
- Regression guard:
- Commands:
Use one-shot Explorer when you cannot find the relevant boundary. Use one-shot Fixer only when the bug spans multiple files and you have a hypothesis plus a failing command.
If the same fix strategy fails twice:
progress.md.BLOCKED.Report symptom, root cause, changed files, regression proof, commands run, gate updates, and unresolved risk.
This skill is a lean alloy-specific overview. When an investigation gets hard, read these files bundled next to this SKILL.md (they ship with the skill):
references/root-cause-tracing.md — tracing a failure to its originreferences/condition-based-waiting.md (+ condition-based-waiting-example.ts) — replacing flaky sleeps with condition waitsreferences/defense-in-depth.md — layering guards so a class of bug cannot recurreferences/find-polluter.sh — bisecting test pollution / order-dependent failuresSource (fusionType inline-append): obra/superpowers v5.1.0 systematic-debugging. Verbatim upstream copies also live under vendor/skills/superpowers/5.1.0/systematic-debugging/ in the Alloy source repo.