Skip to content

AI Security reference architecture

AI agent security

Scoped permissions, approvals and monitoring for autonomous and semi-autonomous agents.

Agent controlsinspection point

Design principles

  • Agents act on behalf of an identity
  • High-impact actions need a human
  • Sandbox code execution
  • Baseline and alert on behavior

Components

ComponentRole
Agent registryApproved agents and owners
Scoped credentialsShort-lived, task-bound access
Approval workflowHuman review for risky actions
Behavior monitoringAnomaly detection

Adapt this to your environment

We'll help you apply this pattern to your constraints and existing estate.