Open Policy Agent

Open Policy Agent (OPA) is an open-source policy engine for evaluating authorization and policy decisions over structured input.

Key Points

• Policies are written in Rego.
• Applications query OPA with structured input and receive policy decisions.
• Used for API authorization, Kubernetes, CI/CD, infrastructure, and other policy surfaces.

Relevance to Init Intelligence

OPA is useful for product experiments around policy-as-code: whether an agent can call a tool, whether a workflow requires approval, and whether a requested action violates tenant policy. ^[inferred]

Related

• agent-tool-governance
• ai-agent-governance-layer
• Engineering index
• Governance stack peers: Arcade AI, Composio, OpenFGA, Cedar, E2B
• research-initlabs-ai-itsm-engineering-stack