Provides action-control workflows and trust enforcement for AI agent systems using Runtime Gate and human-in-the-loop approvals. It enables agents to evaluate risky actions, verify audit evidence, and manage advisory reputation through an M…