← ClaudeAtlas

diagnosing-bugslisted

Diagnose a bug by building a tight failing signal before theorising about the cause. Use when a failure, exception, flaky test, or slowdown has no obvious cause, or the user says "diagnose" or "debug this".
RIJUDEVx/agent-skills · ★ 1 · AI & Automation · score 74
Install: claude install-skill RIJUDEVx/agent-skills
# Diagnosing Bugs A discipline for hard bugs. Skip phases only when explicitly justified. When exploring the codebase, read the relevant domain glossary to get a clear mental model of the modules you touch, and check the ADRs in that area. Find them the way the repo points to them: `.agents/docs/domain.md` when it exists, then `GLOSSARY-MAP.md` → `{product}/GLOSSARY.md` for a multi-context repo, otherwise `GLOSSARY.md` at the root or near the code; ADRs live wherever the repo says, commonly `docs/adr/` or `.knowledge/adr/`. If the repo says nothing, skip them rather than inventing a path. ## Redact This skill has you show commands, outputs and captured artifacts. **Redact every secret first**: write `<REDACTED>` in its place. Build loops against env vars, so the credential stays in the environment rather than in what you show. Captured artifacts carry auth headers: quote only the lines that carry the signal. If the redacted output is not enough to diagnose the bug, say so and ask the user. ## Phase 1: Build a feedback loop **This is the skill.** Everything else is mechanical. If you have a **tight** pass/fail signal for the bug (one that goes red on _this_ bug), you will find the cause; bisection, hypothesis-testing, and instrumentation all just consume it. If you don't have one, no amount of staring at code will save you. Spend disproportionate effort here. **Be aggressive. Be creative. Refuse to give up.** ### Ways to construct one, in roughly this order 1. **Fail