Agents
The AI agents operating your apps — what they’ve paused for, what they asked about, and what looked off. Decide here; Warden records every call in the Audit log.
Approvals
Risky actions an agent asked about first. Each shows what it wants to do, why, and the rule that stopped it.
Nothing pending. Guarded actions land here as agents attempt them.
Anomalies
Allowed actions that broke an agent's pattern. Nothing is enforced until you decide.
No open anomalies. Warden is watching the autonomous stream — drift shows up here.