Autonomy is opt-in. A new organisation starts with auto-remediation disabled and no active agents - nothing runs autonomously until an admin turns it on here.
Global Pause
The pause control at the top of the page is the emergency brake: one action disables the auto-remediation policy and pauses every active AI agent. Resuming restores exactly what the pause switched off - it never enables anything that was off before the pause. The pause record is tracked in the browser that paused, so resume from the same machine where possible.Auto-Remediation Policy
When enabled, the remediation agent can act on detected incidents. The policy controls its reach:- Autonomy level - Observe only (report, never act), Propose (draft actions for approval), or Autonomous (low-risk reversible actions run on their own; the rest ask first).
- Approvals - approval requests go to a Slack webhook; you can name the users allowed to approve.
- Limits - a cap on actions per incident and a cooldown between incidents on the same resource.
- Per-action rules - override individual tools (for example
delete_podordrain_node) to Run automatically, Ask first, or Never run. - Cluster scope - apply the policy to all clusters or an explicit allow-list.