Governance & Compliance · Data artifact

Agent-Specific Policy

Data artifactGovernance & ComplianceSafety, Security & Governancearc:AgentSpecificPolicy

An enforceable policy set scoped to a single agent deployment that handles edge cases and experiments, e.g., reduced transaction limits and mandatory approvals while a new agent is validated.

Responsibility. Imposes per-agent restrictions during edge cases or validation periods.

Also known as: Agent policy tier, Validation-period restrictions

Direct neighbourhood (hover for relationship types)

Relationships

configures structural

Action Policy Engine Ch10.5

constrains control

Agent Controller abstract Ch10.5

Design guidance

SHOULD impose restrictive limits on newly deployed agents (low transaction limits, pre-approved suppliers, approval for all contract changes) and relax them gradually as testing builds confidence.

Classification

Patterns: Risk-based policy tieringGraduated relaxation
Quality attributes: Safety (ISO/IEC 25010 | NIST AI RMF: safe)

Sources

Ch10.5: T. Nguyen, "Human-over-the-Loop," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.5. ISBN: 9798244538229.