Agent Identity, Auth, and Secrets

Two questions decide every agent authorization design: what may THE AGENT do, and on WHOSE BEHALF is it acting? Per-agent identity, on-behalf-of delegation, short-lived task-scoped credentials, secrets that never touch the context window, and audit trails that survive an incident.

Content current as of 2026-09.

Lessons

1. The identity problem: whose power is it wielding?
2. Three patterns: service principal, on-behalf-of, task-scoped
3. What the platforms verifiably ship
4. The context window is not a vault
5. Audit trails: attributable to agent, principal, and task