bug-hunt

Identify and verify reproducible software bugs with severity-based triage.

Updated Apr 3, 2026
One-click install
npx skills add https://github.com/strawgate/agent-skills --skill bug-hunt-strawgate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bug-hunt
Source: https://github.com/strawgate/agent-skills/tree/main/bug-hunt
Command: npx skills add https://github.com/strawgate/agent-skills --skill bug-hunt-strawgate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the challenge of finding and verifying real, user-impacting bugs in a codebase without getting lost in noise, style nits, or speculative issues.

Core Features & Use Cases

  • Systematic Reproduction: Enforces a strict requirement for minimal, reproducible artifacts for every bug found.
  • Multi-Agent Coordination: Provides a structured workflow for multiple agents to claim, hunt, and deduplicate findings in a shared environment.
  • Severity-Based Triage: Facilitates the promotion of candidates into verified critical, high, medium, or low severity reports.

Quick Start

Use the bug-hunt skill to scan the current repository for user-impacting bugs within the authentication subsystem.

Frequently Asked Questions about bug-hunt

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

FAQPage Schema
How do I systematically find and verify reproducible software bugs in a codebase?

Systematic bug hunting isolates real, user-impacting bugs by enforcing strict requirements for minimal reproduction scripts and structured evidence, ensuring findings are deduplicated and verified rather than speculative noise.

What is the best way to isolate state-machine failures and boundary-condition errors?

Isolating state-machine failures and boundary-condition errors requires evidence-driven bug hunting that enforces minimal, reproducible artifacts and adheres to a defined severity-based triage process to verify findings.

How do I triage software bugs by severity without getting lost in style nits or speculative issues?

Severity-based triage promotes candidate bugs into verified critical, high, medium, or low severity reports, filtering out noise and speculative issues by requiring structured evidence for every finding.

Can I use a multi-agent workflow to hunt and deduplicate bugs in a shared environment?

Multi-agent coordination provides a structured workflow for multiple agents to claim, hunt, and deduplicate bug findings in a shared environment, operating across diverse codebases to isolate regressions.

Do I need minimal reproduction scripts to verify software regressions during bug hunting?

Yes, bug hunting requires minimal, reproducible artifacts for every bug found to verify software regressions, ensuring each finding is backed by structured evidence before triage assessment.

When should I not use an evidence-driven bug hunting approach for codebase verification?

Evidence-driven bug hunting is not suited for resolving style nits or speculative issues, as it strictly requires minimal reproduction scripts and structured evidence to verify user-impacting regressions.