Agent identity
The identity an AI agent's actions are attributed to and authorised under. Most agents today act as the developer who launched them, or with a shared key, so logs cannot tell the person from the agent; a distinct, scoped identity for the agent lets access be granted to the agent itself, and its actions be traced and revoked separately.
How MoorAI handles it: Every MoorAI alert names the agent it is about (claude-code, codex, cursor, gemini, copilot, the gateway, the model proxy and others), and alerts from server-side surfaces carry the workload's identity.
See also non-human identity · confused deputy · W3C Trace Context
Related terms
-
Non-human identity
Agents & architecture
An identity used by software rather than a person — a service account, API key, OAuth token, certificate or workload identity — usually backed by a secret…
-
Confused deputy
Attack concepts
A classic security problem, dated 1988, that agents reproduce almost perfectly: a privileged component is tricked by a less-privileged party into misusing…
-
W3C Trace Context
Frameworks & standards
A W3C Recommendation (November 2021) that standardises the HTTP headers carrying distributed-tracing context between services: traceparent gives a…
Agents & architecture
Where the attacks above actually land. The vocabulary here is worth precision, because the difference between "the model decided" and "the tool ran" is the difference between a risk and an incident.