I Built a CLAUDE.md for Bug Bounty. Here's What's In It.

I've been using Claude Code for almost every hunt lately; but out of the box, it doesn't think like a bug bounty hunter. So, I built a system to fix that. A CLAUDE.md that primes the mindset. And a set of custom skills that give it the tools to act on it. In this video I'm breaking down the whole setup. Every file, what's in it, and why I built it the way I did. What's covered: → CLAUDE.md: how I prime Claude to approach a new target, recon methodology, output formatting, and OpSec rules → triager.md: simulates a real Bugcrowd triager before anything gets submitted. Brutal. Honest. Protects the account. → report-draft.md: structures findings into clean, submission-ready reports → hypothesis.md: generates attack hypotheses for a given target surface All files are live on GitHub. Grab them, break them, rebuild them your way. github.com/logan-sec/claude-code-bb Get the free JS Recon Automation Kit: https://forms.gle/RsGu6qoCJBGFjQjL8 I’m releasing the script to the private email list first. It turns a HAR file into organized JS files, LinkFinder endpoints, SecretFinder hits, and gf-filtered attack surface for manual bug bounty review. — 𝕏 @LoganOpSec GitHub: github.com/logan-sec #bugbounty #cybersecurity #claude #ethicalhacking