# Debugging Approach

> Applies root-cause isolation, binary search debugging, error trace analysis, and systematic troubleshooting methods.

- Skill: `felipecustodio/debugging-approach` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add felipecustodio/debugging-approach`
- Raw SKILL.md: https://api.skillmd.com/api/skills/felipecustodio/debugging-approach/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: felipecustodio (https://skillmd.com/u/felipecustodio)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/felipecustodio/debugging-approach

---


## Overview & Purpose

Debugging Approach is a systematic method for isolating software bugs, performance bottlenecks, and hardware/network errors without reliance on guess-and-check.

## When to Use

- **Software Debugging**: Investigating stack traces, test failures, memory leaks, or unexpected output.
- **System Troubleshooting**: Diagnosing configuration errors or environment mismatches.

## Execution Workflow

1. **Reproduce & Isolate**: Establish minimal reproducible example.
2. **create Hypotheses**: List potential failure points based on error signature.
3. **Binary Search Isolation**: Divide search space in half (e.g., git bisect, narrowing down line numbers or components).
4. **Inspect State**: Verify actual vs expected values at boundaries.
5. **Fix & Verify**: Apply minimal necessary fix and verify no regressions.

## Expected Output Contract

```markdown
### Debugging Report
- **Error Signature**: [Stack trace / symptom]
- **Isolation Boundary**: [Component / line identified]
- **Root Cause**: [Mechanism of failure]
- **Verification**: [Test confirming fix]
```

## Scripts

- `scripts/debugging_approach.py` - Deterministic evaluation, state validation, and CLI tool for debugging-approach.