| name | fix-bugs |
| description | Autonomously fix bugs from any source — Slack threads, failing CI tests, docker logs, or a plain description. Use when you have a bug report or failing test and want Claude to investigate and fix without step-by-step guidance. |
| license | MIT |
| compatibility | Designed for Claude Code. Works best with access to CI logs, Slack MCP, or docker CLI. |
| metadata | {"author":"rajit","version":"1.0"} |
| allowed-tools | Bash Read Grep Glob Edit Write |
Fix the bug described or linked. Do not wait for step-by-step instructions — investigate and fix autonomously.
Input Sources (handle any of these)
- Slack thread URL or paste — Read the thread, extract the bug report, reproduce and fix
- "Go fix the failing CI tests" — Run or read CI output, identify failures, fix them
- Docker logs — Parse logs for errors/stack traces, trace to root cause, fix
- Plain description — Understand the bug, find the relevant code, fix it
Process
- Gather context — Read the error message, stack trace, or thread in full. Do not skip this.
- Locate the code — Find the exact file(s) and line(s) responsible. Use Grep/Glob aggressively.
- Understand before fixing — Read the surrounding code. Do not patch symptoms.
- Fix — Make the minimal correct change. Do not refactor unrelated code.
- Verify — Run tests or check that the specific failure condition is resolved.
- Summarize — One sentence: what was wrong and what you changed.
Principles
- Do not micromanage the approach — just say "fix" and trust the process
- Point at docker logs for distributed system issues — Claude can trace across service boundaries
- If CI is failing: read the test output first, then the test code, then the implementation
- Do not add workarounds or feature flags — fix the root cause
- If you cannot reproduce or find the root cause, say so clearly rather than guessing
Examples
User: fix this https://slack.com/archives/C07VBSH.../p1234
→ Claude reads Slack thread, finds the bug, fixes it
User: go fix the failing CI tests
→ Claude reads CI logs/test output, traces failures, patches code, reruns
User: look at these docker logs and fix whatever's broken
[paste logs]
→ Claude parses errors, finds the service + file, fixes it
