Skip to content

Claude Hook Policy Simulator

Last updated:

This tool is provided as-is for convenience. Output should be verified before use in any production or critical context.

Agent Invocation

Best Path For Builders

Browser workflow

Runs instantly in the browser with private local processing and copy/export-ready output.

Browser Workflow

This tool is optimized for instant in-browser execution with local data handling. Run it here and copy/export the output directly.

/claude-hook-policy-simulator/

For automation planning, fetch the canonical contract at /api/tool/claude-hook-policy-simulator.json.

How to Use Claude Hook Policy Simulator

  1. 1

    Define default action and rules

    Paste JSON with defaultAction and ordered hook rules. Each rule can match by phase, tool, and optional command substring.

  2. 2

    Add representative hook events

    Include realistic pre and post tool events from your workflow so the simulation reflects real runtime conditions.

  3. 3

    Run policy simulation

    Execute Run Hook Simulation to evaluate each event against rule order and observe final allow, deny, or ask actions.

  4. 4

    Refine rules before production rollout

    Use decision summaries and unmatched cases to tighten policy logic, then repeat simulation until outcomes match your governance requirements.

Frequently Asked Questions

What is Claude Hook Policy Simulator?
Claude Hook Policy Simulator replays pre and post tool events against your hook rules. It shows what would be allowed, denied, or escalated before you enforce policies in production.
How do I model hook rules?
Define a default action and add ordered rules with phase, tool matcher, optional command text matcher, and target action. Then include sample events to simulate real runtime behavior.
Does it support wildcard policies?
Yes. You can use wildcard values for phase or tool to model broad policy fallbacks and then layer specific exceptions above them.
Does Claude Hook Policy Simulator store or send my data?
No. Simulation runs entirely in the browser and your policy payload stays local.