afc:debug

/afc:debug — Bug Diagnosis and Fix

Analyzes the root cause of a bug and fixes it. Validates the safety and accuracy of the fix with convergence-based Critic Loop.

Arguments

• $ARGUMENTS — (required) Bug description, error message, or reproduction steps

Project Config (auto-loaded)

!cat .claude/afc.config.md 2>/dev/null || echo "[CONFIG NOT FOUND] .claude/afc.config.md not found. Create it with /afc:init."

Execution Steps

1. Gather Information

1. Extract from $ARGUMENTS:

   • Symptom: what is going wrong?
   • Reproduction conditions: when does it occur?
   • Error message: full text if available
   • Expected behavior: what should happen?

2. Ask user for additional information if needed (max 2 questions)

2. Root Cause Analysis (RCA)

Proceed in order, applying adaptive skipping based on symptom type:

1. Error trace: extract file:line from error message/stack trace → read that code
2. Data flow: trace backwards from the problem point (where did the bad data come in?) — skip if error trace already identifies root cause with high confidence
3. State analysis: check relevant state management cache state (from Project Context) — skip if bug is clearly a syntax/type/null-reference error with no state involvement
4. Recent changes: check recent changes with git log --oneline -10 -- {related files} — skip if bug is clearly pre-existing (user reports it "always" fails)
5. Race conditions: check for timing issues between async operations — skip if bug is deterministic (same input always fails, no async/concurrent context)

When skipping a step, note: "Skipped: {reason}". Do not skip steps 1 and 2 unless step 1 alone identifies the root cause.

3. Form Hypotheses

List possible causes as a hypothesis list:

### Hypotheses
0. **[Not a Bug?]** {intended behavior explanation}: {evidence from docs/code/spec}
1. **[High probability]** {cause1}: {evidence}
2. **[Medium probability]** {cause2}: {evidence}
3. **[Low probability]** {cause3}: {evidence}