✦ Agent Skill

Find, verify and report bugs within authorized scope

Within authorized scope, have your AI agent hunt for vulnerabilities, verify them, and write the report using a disciplined workflow.

Author CWYZLevel Intermediate
4.0
Free

Use this when you want to run a security assessment within scope you own or have written authorization for (reviewing your own project, an authorized penetration test, a CTF challenge, or hardening against the OWASP checklist), and want your AI agent to follow a disciplined, reproducible staged process that produces evidence and a report. Based on the open-source Claude-BugHunter (MIT / CC BY 4.0): confirm the authorized scope first, then map the exposed surface, check it against a checklist, verify each finding one by one, and write the report in the platform's format. Test only authorized targets. No post-exploitation or attack-infrastructure development.

Updated Nov 16, 2026

Skill name

claude-bughunter

Works with

Claude CodeCodexAny agent that reads SKILL.md

Access

Free after sign-in

Similar skills

All skills

What you get

  • A reproducible vulnerability report with evidence
  • A verification of whether each finding is real
  • A step-by-step assessment process

Good for

  • Developers reviewing their own projects
  • People doing authorized penetration testing or CTFs

Not for

  • Targets without written authorization
  • People who only want to run automatic scanners

About this skill

Lets your AI agent run a disciplined security assessment within authorized scope. It first confirms the scope of authorization, then maps the exposed surface, checks against a checklist, verifies each finding is real, and finally writes a report in the platform's format.

What you need

  • An AI agent that supports skills. If it supports plugins, install through the plugin. Otherwise, clone with git and run the install script included in the repo.
  • A target you have permission to test: your own assets, a target in scope of a bug bounty program, a penetration test with written authorization, or a CTF challenge
  • The skill itself does not need an API key
  • The clone also brings a command-line tool and task scaffolding. Before installing, have the AI agent read the repo's install instructions.

FAQ

Can I use it without written authorization?

No. It is only for targets you own or have written permission to test. Without authorization, it stops right away.

Will it help with post-exploitation or evading detection?

No. Persistence, lateral movement and evading detection are not part of this workflow.

What if the AI agent refuses to continue halfway?

Authorized offensive testing can hit the model's safety limits. The right way is to apply for the proper access through the official instructions, not to rephrase the task to get around it.

Will it report lots of false positives?

Every suspected issue must be verified first. Anything that can't be confirmed won't be written up as a confirmed vulnerability.

This is an outline. The full SKILL.md comes with the install: sign in, then copy the prompt or download.

REVIEWS

4.0

0 reviews

Rate this skill

No reviews yet.

Used this skill? Be the first to rate it.

Done with this one? Mark it to track your path.Sign in
Save it for laterSign in
Questions about this piece?Ask AI about this