Skip to content
Get packs

Hunt: Root-Cause Debugging

Makes the agent find and prove a bug's root cause before changing any code.

Source Download skill
Repository
tw93/Waza
Stars
7,493
Licence
MIT
Files
6 files

What it does

When something errors, crashes, regresses, or stops working, this skill makes the agent stop and form a specific, testable explanation of the cause before touching any code. It asks the agent to check every reported symptom against that explanation, reproduce the problem, read logs and state, and only then apply a fix that is verified by a repeatable check. It also sweeps the codebase for the same bug pattern elsewhere and reports what it finds.

Great for

  • Developers chasing a crash, error, or test failure that keeps coming back after quick patches
  • Teams investigating a regression where something worked in an earlier commit or version
  • Anyone whose agent keeps applying fixes that don't stick and needs a more disciplined process

Skip it if

  • Code review or designing new features, which the skill explicitly excludes (the agent may do a poor job at these using a debugging workflow)
  • Purely subjective visual taste questions, which the skill says should go to a different skill (/ui)
  • Bisect-style investigations in a repository with uncommitted changes: the skill requires a temporary worktree or explicit user approval, so careless use could still touch the working copy

Try asking

  • “Why is the login page throwing a 500 error after yesterday's deploy? Find the root cause first, don't just patch it.”
  • “This used to work before the last update. Figure out which commit broke it.”
  • “The button color is still wrong after your last fix, here is the screenshot that looked right before. Investigate.”
  • “Find out whether the same null-check bug exists in other handlers in this repo.”
Step-by-step guide: how to use Hunt: Root-Cause DebuggingInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 6 original files, the MIT licence and an attribution note.

    hunt.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip hunt.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip hunt.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Say 'diagnose' or 'investigate' if you want a report only; the skill will not apply a fix unless you explicitly ask for one.
  • Give the agent a known-good reference (a commit, build, or screenshot) and a clear description of the wrong behavior to speed up diagnosis.
  • Ask for the blast-radius sweep explicitly if you want other places with the same bug pattern checked.

What's inside

  • references/durable-context.md
  • references/failure-patterns.md
  • references/ime-unicode.md
  • references/logging-techniques.md
  • references/rendering-debug.md
  • SKILL.md

Redistributed unmodified from tw93/Waza under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.