Talk to us
Rego · OPA v1

An AI agent policy template for OPA

A policy engine is only useful if the model cannot argue with it. This Rego starter decides one question per tool call — allow, require a named human, or deny — from the tool name, the arguments and the caller’s scopes, and it fails closed on anything it does not recognise.

agent-action-gate.rego No signup, no email.

It is written as a starting point, not a product. The eight decision cases in the file are the specification: read them first, then adapt the thresholds to the workflow you are governing.

What is inside

SectionWhat it covers
Decision shapeallow / deny / requires-approval, with a reason string the runtime can log and show a reviewer.
Fail-closed defaultsOnly a read-only tool the agent holds a scope for is auto-allowed. Everything unknown is denied.
Arguments, not namesAmounts over a limit, personal data classes and write operations route to a named human rather than passing.
Test fileagent-action-gate_test.rego covering the eight decision cases documented in the policy header.

How to use it

  1. Run opa test on the two files before you change anything. If the baseline fails, the OPA version differs from the one this was written against.
  2. Replace the sample tool names and scopes with your own registry entries.
  3. Set the spend and data-class thresholds to the values your risk owner will defend to an examiner.
  4. Deploy in log-only mode first for a week and compare what the policy would have blocked with what actually happened.
  5. Move to enforce once the denial list is boring. Keep the last versions so a change can be explained.

How do you write a policy that decides what an AI agent may do?

Does this policy work with Open Policy Agent v1?

Yes — it is written for OPA v1 syntax. The eight decision cases were checked on an independent Rego v1 engine on 2026-10-03; run `opa test` against your own OPA binary before you rely on it, since the test file and the binary have not been run together here.

Why fail closed rather than warn?

A warning arrives after the call has already reached the system of record. Fail-closed means the default answer to an unrecognised tool, an unscoped caller or an oversized amount is no, and the reviewer sees the denial instead of a customer seeing the mistake.

Can I use this commercially?

The file is published as Apache-2.0 (proposed in the file header). Adapt it freely; keep the attribution notice if you redistribute it.

Where this fits