Recovery
Recovery & emergency
Stopping is always available and never gated. These guides cover stopping quickly, undoing what can be undone, recognising what cannot, and returning to normal without leaving the cause in place.
- 6 guides
- 23 min to read all
- Recovery
Guides in this collection
- 4 min readUse read-only and paused
Three site modes that take effect immediately across every agent. Which to reach for, what each still permits, and why stopping is never gated by plan.
- 3 min readStop a single agent immediately
The fastest way to stop one agent without stopping the others, and the order to do things in so nothing slips through while you work.
- 5 min readUndo an unwanted change
Work from the record: stop the source, identify what changed, use WordPress’ own recovery paths, then narrow the permission before resuming.
- 4 min readWhen an action cannot be undone
Reversibility is a property of the action, known in advance from its classification. What to do when there is no undo, and how to avoid a repeat.
- 3 min readTerminate sessions and credentials
Sessions are ended connection-wide and credentials invalidated before the connection state changes, so nothing in flight outlives the decision.
- 4 min readReturn to normal after an incident
A checklist for coming out of an emergency mode: change the cause first, verify the controls, then restore service — and write down what you changed.
Other collections
- 6 guidesInstall & activate
Requirements, activation, updates, and first-run checks.
- 8 guidesAgents & identity
Recognition, environment, status, and managed identity.
- 12 guidesAbilities & permissions
Registration, native authority, scope, and policy.
- 7 guidesApprovals
Review windows, request binding, expiry, and replay refusal.
- 9 guidesActivity & evidence
Decision records, verification, redaction, and export.
Still stuck?
Bring the request, not just the symptom.
Identity, Ability, decision, outcome, evidence — then the question.