Is the Bug Hunt skill free?

Yes. The skill itself is free from chrisallenlane/claude-swe-workflows. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Bug Hunt SkillProof Score?

9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.

How do I install Bug Hunt?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Bug Hunt with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Bug Hunt

Proactive bug hunt that reports only findings backed by a failing test

by chrisallenlane · chrisallenlane/claude-swe-workflows

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 9.2/10
Tested: Aug 7, 2026
Environment: Claude Code 2.x (agent harness)

Pointed it at a third-party Python module with zero existing tests: reading the code alone produced nine plausible-sounding bug claims, and the skill's rule that every finding needs a failing test killed two of them outright while upgrading three to hard evidence. The strongest was a crafted metric name that made the query return the row count of the supplements table instead of daily calories — a working SQL injection, not a suspicion. The two hypotheses that survived reading but not execution are exactly the noise this workflow is built to delete.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 8/10
Docs & honesty 5/5

What Bug Hunt does

Ranks a codebase by converging complexity, coverage and structural-risk signals, then sends focused hunters into each hotspot to write a reproducing test for every suspicion and keep only the ones that fail. Triggers on pre-release quality passes, post-refactor sweeps, and getting oriented in an unfamiliar codebase. Advisory only: it cuts tickets with the failing tests as acceptance criteria and never implements the fix.

How to install Bug Hunt

git clone https://github.com/chrisallenlane/claude-swe-workflows.git
mkdir -p ~/.claude/skills ~/.claude/agents
cp -r claude-swe-workflows/skills/bug-hunt ~/.claude/skills/bug-hunt
cp -r claude-swe-workflows/references ~/.claude/references
cp claude-swe-workflows/agents/*.md ~/.claude/agents/

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Bug Hunt

/bug-hunt Proactive bug hunt that reports only findings backed by a failing test

It also activates on plain-language prompts like these:

Hunt for bugs in this codebase and back each one with a failing test
Do a pre-release bug sweep on this module before we ship it
Find real bugs in this refactored code, only report ones with proof

Frequently asked questions

Is the Bug Hunt skill free?: Yes. The skill itself is free from chrisallenlane/claude-swe-workflows. SkillProof publishes the install command and an independent test verdict at no cost.
Does Bug Hunt work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Aug 7, 2026. Verdict: Tested · Works. Pointed it at a third-party Python module with zero existing tests: reading the code alone produced nine plausible-sounding bug claims, and the skill's rule that every finding needs a failing test killed two of them outright while upgrading three to hard evidence. The strongest was a crafted metric name that made the query return the row count of the supplements table instead of daily calories — a working SQL injection, not a suspicion. The two hypotheses that survived reading but not execution are exactly the noise this workflow is built to delete.
What is the Bug Hunt SkillProof Score?: 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
How do I install Bug Hunt?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Bug Hunt with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.