Debugging

Systematic approach to root cause analysis and debugging.

When to Use

• Encountering errors or exceptions
• Test failures that need investigation
• Unexpected behavior in code
• Stack traces or error messages
• Code behaving differently than expected
• Performance issues or bugs

Core Principles

• Evidence-based: Base diagnosis on error messages, logs, and reproducible steps
• Systematic: Follow structured debugging process
• Minimal fixes: Implement smallest change that resolves issue
• Verify solutions: Confirm fix works and doesn't introduce regressions

Debugging Process

Follow systematic debugging process:

1. Capture error information (message, stack trace, logs, environment)
2. Identify reproduction steps (minimal steps, conditions, edge cases)
3. Isolate failure location (function/module, recent changes, dependencies)
4. Form and test hypotheses (evidence-based, systematic testing, debug logging)
5. Implement minimal fix (smallest change, preserve behavior, follow patterns)
6. Verify solution (issue resolved, no regressions, tests pass)

See references/root-cause-analysis.md for detailed methods.

Strategic Debug Logging

Add debug logging to entry/exit points, state transitions, conditional branches, external API calls, and data transformations. Remove after issue resolved unless it provides ongoing value.

Error Pattern Recognition

Common patterns: null/undefined errors, type errors, timing issues, state corruption, configuration issues. See references/error-patterns.md for detailed patterns and solutions.

Integration

After fixing:

• Verify CI passes (types, tests, lint)
• Stage atomic changes (fix + tests)
• Suggest semantic commit message
• Confirm with user before committing

References

For detailed guidance, see:

• references/root-cause-analysis.md - Systematic analysis methods
• references/error-patterns.md - Common error patterns and solutions
• references/debugging-tools.md - Debugging tools and techniques