The agent prompt
Paste one prompt into Claude Code or another coding agent. It installs the skill, audits your repo and sums up the report.
Cloudflare open sourced the skill its own vulnerability hunting started with. This guide installs it in your coding agent with one command and runs your first full security audit of your app from a single sentence.
Paste one prompt into Claude Code or another coding agent. It installs the skill, audits your repo and sums up the report.
What the skill checks, and why a second agent tries to disprove every finding before it reaches your report.
Which of the three files to open first, and what confirmed, needs validation and rejected mean.
π Full guide after signupYes. You sign up with your email once and get this guide plus every other guide in the library.
Yes. Cloudflare published it as open source under the MIT license. You only pay for the AI agent or model you already use to run it.
Only inside a locked-down sandbox with no network. If your setup cannot enforce that, the skill reads the source only and marks the open questions as needs validation instead of running anything.
We send a confirmation email. Click the link and the full guide opens. You also get the ai-toolfinder newsletter every Friday and can unsubscribe anytime.