Skill1.3k repo starsupdated today

structured-debug

The structured-debug skill guides users through a systematic debugging workflow before modifying code. It emphasizes clarifying expected versus actual behavior, reproducing issues with environmental details, testing hypotheses with specific evidence checks, planning fixes with risk assessment, and validating solutions against regression. Use this skill when addressing bugs, investigating regressions, triaging incidents, diagnosing failing tests, or performing root cause analysis to ensure thorough problem-solving before code changes.

View source Repository: ai-devkit

Install in Claude Code

Copy

git clone --depth 1 https://github.com/codeaholicguy/ai-devkit /tmp/structured-debug && cp -r /tmp/structured-debug/skills/structured-debug ~/.claude/skills/structured-debug

Then start a new Claude Code session; the skill loads automatically.

Definition

SKILL.md

# Local Debugging Assistant

Debug with an evidence-first workflow before changing code.

## Hard Rule
- Do not modify code until the user approves a selected fix plan.

## Workflow

1. Clarify
- Restate observed vs expected behavior in one concise diff.
- Confirm scope and measurable success criteria.
- Before investigating, search for similar past incidents: `npx ai-devkit@latest memory search --query "<observed behavior>" --tags "debug,root-cause"`

2. Reproduce
- Capture minimal reproduction steps.
- Capture environment fingerprint: runtime, versions, config flags, data sample, and platform.

3. Hypothesize and Test
For each hypothesis, include:
- Predicted evidence if true.
- Disconfirming evidence if false.
- Exact test command or check.
- Prefer one-variable-at-a-time tests.

4. Plan
- Present fix options with risks and verification steps.
- Recommend one option and request approval.

## Validation
- Confirm a pre-fix failing signal exists.
- Confirm post-fix success using the `verify` skill — including regression verification for bug fixes.
- Summarize remaining risks and follow-ups.
- Store root cause and fix for future sessions: `npx ai-devkit@latest memory store --title "<root cause>" --content "<diagnosis and fix>" --tags "debug,root-cause"`

## Red Flags and Rationalizations

| Rationalization | Why It's Wrong | Do Instead |
|---|---|---|
| "I already know the cause" | Assumptions skip evidence | Reproduce and prove it first |
| "This is urgent, just fix it" | A wrong fix wastes more time | 10 minutes of diagnosis saves hours |
| "The fix is obvious from the stack trace" | Stack traces show symptoms, not causes | Trace backward to the root cause |

## Output Template
Use this response structure:
- Observed vs Expected
- Repro and Environment
- Hypotheses and Tests
- Options and Recommendation
- Validation Plan and Results
- Open Questions