What this map covers
A map of the control mechanisms around agent actions.
My current view: governance only works when permission boundaries, approvals, and rollback paths are explicit.
Core concepts
- permission-boundary
- approval-gate
- audit-trail
- rollback
- write-access
- process-orchestrator
- agent-trace
- agent-evaluation
- observability
Important claims
Field notes
Open questions
- Where do we enforce policy, and where do we only observe?