Menu
Systematic debugging - find root cause before attempting fixes
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Automatically implement all sub-issues of an epic in dependency order
Implement a GitHub issue with automated PR creation
End-of-session retrospective. Captures knowledge from any working session (debug, implementation, config, deployment) as reusable scripts, CLAUDE.md procedures, or skill proposals. Run before ending a session to prevent knowledge loss.
Decompose a large GitHub issue into sub-issues and create a tracking draft PR
OWASP-guided security code review for a specific domain or issue
Project kickoff checklist based on lessons learned from previous projects. Generates a tailored checklist and optionally creates GitHub issues.
| name | debug |
| description | Systematic debugging - find root cause before attempting fixes |
| argument-hint | <description of the bug or error> |
| user-invocable | true |
Find the root cause before attempting any fix. Random fixes waste time and create new bugs.
The user describes a bug, error, or unexpected behavior: $ARGUMENTS
NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST
If you haven't completed Phase 1, you CANNOT propose fixes.
BEFORE attempting ANY fix:
Show the error — present the full error message and stack trace to the user
Reproduce consistently
Check recent changes — git diff and recent commits, new dependencies, config changes
Trace data flow — where does the bad value originate? Trace backward to the source
For multi-component systems: add diagnostics first — log what enters/exits each component boundary, run once to find WHERE it breaks
If you've tried 3 fixes and none worked:
STOP. Do not attempt fix #4.
This pattern indicates an architectural problem, not a bug:
Discuss with the user before continuing. This is not a failed hypothesis — this is a wrong architecture.