Skip to content
navigateopenescclose
Claude+372Whisper+228LangChain+168Codex+223NotebookLM+276DALL-E 3+192DeepL+249n8n+208Topaz Video AI+153LlamaIndex+161
Debugging Toolkit logo
Skill

Debugging Toolkit

Ranks debugging hypotheses and gives each one a test that would prove it wrong.

0
SaveVisit website ↗

Install

Claude Code: /plugin marketplace add wshobson/agents → /plugin install debugging-toolkit
Triggers on

Bir hata ayıklanırken, üretimdeki arızanın kök nedeni aranırken ya da geliştirme ortamının sürtünmesi azaltılırken devreye giriyor.

Needs tools
ReadBashGrep
Author
Seth Hobson
License
MIT

The smart-debug command triages an error, collects whatever observability the project already has — error tracker, APM metrics, distributed traces, aggregated logs — and produces three to five ranked hypotheses. Each carries a probability, the evidence behind it and, less usually, the criterion that would falsify it; that last field is what stops a session from circling the first plausible story for an hour. Failures are correlated against the deployment timeline, so configuration drift and a bad release stop being mistaken for logic bugs. A debugger agent does the reproduction and narrowing work, and a separate DX agent attacks the friction around it: time from clone to running app, git hooks, slow builds and error messages that send people to the wrong file.

Comments(0)

Sign in to comment

No comments yet — be the first.

Similar skills

Report this comment

Why are you reporting this?