responsible-disclosure
Automated LLM red-teaming framework. 1,028 attack techniques, reasoning-layer attacks, and five compliance scorecards (OWASP, MITRE ATLAS, EU AI Act). Vendor-neutral, GPL-3.0.