Agent Security Governance

A security management framework for AI agents operating in production environments, encompassing mechanisms such as logging, operation allowlisting, anomalous behavior detection, and human review. Its purpose is to prevent agents from executing unauthorized or dangerous operations.