Identity

Agents & identity

A managed identity answers the first question every governed request asks: who is acting? These guides cover creating agents, connecting them, moving them between states, and what to do when one stops being recognised.

8 guides
31 min to read all
Identity

Guides in this collection

4 min readCreate your first managed agent
Name an agent for the work it does, tie it to one environment, and see why one agent per purpose is what produces a readable audit trail.
5 min readAgent not recognized
The request arrived without a resolvable managed identity, so it stopped at the first gate. Check state, environment, connection and credential.
5 min readChoose an agent’s connection user
The connection user’s role is the ceiling for everything the agent can ever do. How to pick a least-privilege user, and why your own account is not it.
4 min readConnect an agent with the wizard
Generate an Application Password for a dedicated user, copy it once, and confirm the connection is healthy before handing it to an agent.
3 min readRotate a credential without losing history
Rotation replaces the secret while keeping the agent, its permissions, and its audit trail intact. What breaks briefly, and what never does.
4 min readAgent states: which one to use
Agent state is a control surface, not a status label. What each state refuses, and which one to reach for in an incident.
3 min readKeep staging and production agents separate
Environment is part of an agent’s identity. Why the same agent name on two sites is two records, and how to promote a boundary deliberately.
3 min readRevoke an agent and end its sessions
Revocation invalidates credentials before the connection state changes and terminates sessions connection-wide, so nothing in flight outlives the decision.

Other collections

Still stuck?

Bring the request, not just the symptom.

Identity, Ability, decision, outcome, evidence — then the question.

Open support All collections