Red Team

Attacker's-eye test of LLM/agent defenses: instruction hijacking, data exfiltration and tool abuse through untrusted content; verifies whether the defense actually holds.

byerlikaya Updated 26 repo stars
Command flow: /crew-plan, the specialists, /crew-review, /crew-ship with your approval, /crew-handoff

File contents

byerlikaya/claude-starter-kit commit dac8d98ef1

Frequently asked questions

npx get-skill add byerlikaya/red-team-3