Sandbox · Training Mode · no production data, no install

Build an authorize call in your browser.

Pick the actor, principal, capability and scope. The page builds the request, shows the JSON, prints a curl you can copy, and displays a deterministic stub Decision Map™ response in the same shape the reference Gateway returns.

This sandbox runs entirely in your browser. The response is a deterministic stub — for live authorize calls, run the reference Gateway locally (npx wrangler dev (in internal)) and replace the URL.

rocket_launch Apply for pilot → menu_book Read the docs

Authorize playground.

Every claim here is backed by the open KYE Protocol™ contracts and verifiable end-to-end from the publisher's JWKS — you check it yourself, you don't take our word for it.


        

        

          

          

        

        

          

          

        

      

    


    
Schema reference: runtime-authorize-request.json · policy-decision.json.

  






  

    
When you're ready to go live.

      
You see the timing rules that bind every governed decision — admission, replay window, retention floor — all signed and replayable offline.

    
    
      
  1. Run the reference Gateway. npx wrangler dev (in internal). POST to http://localhost:4000/v1/runtime/authorize with the same payload above — you'll get a real signed Decision Map™ back.
    2. 
      
  2. Run the conformance pack. 133 black-box scenarios, byte-for-byte the same as the reference Gateway. See the conformance-run playbook for clone + run + sign + register.
    3. 
      
  3. Wire it into your stack. See the developer portal for SDK quickstarts (TypeScript / Python / Go) and webhook verification.
    4. 
    

  







  

    
Ready to see your AI agents flagged?

    
Start in shadow mode. We’ll deliver your first Evidence Pack™ in 4–8 weeks.

    

      rocket_launch Apply for pilot →
      science Try the sandbox →