hunt-bugs

Generate bug hypotheses, test plans, and PR-ready recovery files for a named subsystem.

Updated Apr 14, 2026
One-click install
npx skills add https://github.com/patrob/skills --skill hunt-bugs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hunt-bugs
Source: https://github.com/patrob/skills/tree/main/plugins/claude-skills/skills/hunt-bugs
Command: npx skills add https://github.com/patrob/skills --skill hunt-bugs

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Proactively discover and verify bugs across a named subsystem by generating hypotheses, writing tests, and guiding minimal fixes that culminate in a ready PR.

Core Features & Use Cases

  • Recon phase to identify 5-10 concrete bug hypotheses with evidence and impact.
  • Parallel test-writers that produce failing integration tests for each hypothesis.
  • Sequential fixers that apply minimal production changes and re-run targeted tests.
  • PR drafting and runbook compilation to document findings and fixes.
  • End-to-end workflow orchestration across Recon, Test Writers, Fixers, and PR stages.
  • Customizable flows per project with artifact aggregation under .pipeline.

Quick Start

Start an automated bug-hunt flow for a given subsystem with /hunt-bugs auto {subsystem}.

Frequently Asked Questions about hunt-bugs

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate bug hunting and generate integration tests for a specific subsystem?

Automated bug hunting identifies a subsystem's boundaries to generate 5-10 bug hypotheses with evidence, then uses parallel test-writers to produce failing integration tests for each hypothesis before applying fixes.

What is the best way to find bugs proactively across a large codebase and prepare a PR draft?

Proactive bug discovery at scale maps a subsystem's entry points, generates concrete bug hypotheses, and orchestrates sequential fixers to apply minimal changes, culminating in a PR-ready recovery plan stored under .pipeline.

How does structured bug hypothesis testing work for integration test generation?

Structured bug hypothesis testing works by mapping subsystem boundaries, creating predetermined test plans for each hypothesis, and executing a sequence of targeted tests and minimal fixes via downstream agents.

Can I use automated test-writers and fixers to create a PR-ready recovery plan for my subsystem?

Yes, you can use parallel test-writers to produce failing tests and sequential fixers to apply minimal production changes, which aggregates artifacts and compiles a runbook into a PR-ready recovery plan.

Does this bug hunting workflow require specific dependencies to run subsystem boundary mapping?

No specific dependencies are required to run subsystem boundary mapping; the workflow operates autonomously to identify entry points, generate hypotheses, and orchestrate the test and fix sequence.

When should I not use a proactive bug hunting approach for my integration tests?

You should avoid proactive bug hunting when you lack a clearly named subsystem to target, as the process requires defining specific boundaries and entry points to generate actionable bug hypotheses and test plans.