Demo. Sample data from a real deployment. Anything you change here lives in this browser session only and is not saved anywhere; sign out to start again.

Krynix Meridian Labs
Enforcement — inferred from observed decision evidence

A policy is evaluated before a call runs, and answers whether it may proceed. A control reads the record afterwards.

To change an existing policy, open it and register a new version — a registered version never changes, so the evidence decided under it stays reproducible.

Baseline policies

A baseline judges every call in the org, whether or not the agent is registered or has its own policies; the strictest verdict wins. To leave an agent out, the policy’s own scope says so.

None. Each agent's calls are decided only by its own policies.

  • coding-agent-guardrails

    1 version · most recently registered 2

    Open policy
    Registered versions of coding-agent-guardrails, most recently registered first
    VersionContent hashRegistered
    2 5b659aa967c131b00507ec25f80576efaf81178a8f1b35f2afb102ba2c64a184
  • deny-all-star

    1 version · most recently registered 1

    Open policy
    Registered versions of deny-all-star, most recently registered first
    VersionContent hashRegistered
    1 372386e5640cac62ddae3f123707fc2e39426f5577c7b052097a9358cc575c39

2 policies and 2 versions registered. Versions are listed in the order they were registered, not sorted — a version is a string the author chose, so 10 after 9 is not something this screen can assume.