A coding-agent skill that turns your agent into a security auditor. It orchestrates isolated agents through reconnaissance, coverage-led hunting, candidate validation, structured output, independent record verification, and target-neutral reporting.
This is the skill that seeded Cloudflare's vulnerability discovery harness, described in Build your own vulnerability harness. The harness grew into a multi-stage, fleet-wide system; this skill is the single-repo starting point it evolved from.
Did you test it?
Not yet, but the things it looks for looked nice.
I'll try to find something small to run it against.
Maybe coldcard firmware before the fix is a nice experiment. Or swiss bitcoin pay mobile app.
Yeah good idea. I have some current apps I have manually reviewed with LLM running inventory and structure discovery only, so I was thinking of doing one of these because I fully know them, but many of these are in a disclosure process. But if it is really good, and it gets new findings, then there's the same issue.