Enables AI agents to evaluate clinical AI models for adversarial robustness, privacy leakage, and standards compliance, producing a Security Posture Score.
