Playground - Skybridge

Putting a real model in the loop normally means registering your app in a host. The Playground skips that: a chat wired to your local server, where a real LLM uses your tools and the views render inline.

Start the Playground

The Playground comes with the tunnel. Start it with the --tunnel flag (or toggle Tunnel in the DevTools header):

npm run dev -- --tunnel

Your local server becomes publicly available, and so does the Playground: it runs on the tunnel URL suffixed with /try (e.g., https://cool-marmot-fondue-420.alpic.dev/try). Anyone with the link can open it from any browser and chat with your app.

Exercise the Model Loop

With a model in the loop, the prompt surface and the state sync become testable:

Tool selection: assess what the model understands from your tool names and descriptions.
Argument filling: check how it fills inputs from your schemas.
Narration: see how it presents your tool outputs in the conversation.
State: verify what it picks up from the view state on its next turn.

Limitations

Host specifics (model behavior, state push policies, view feature support) still need a real host: connect your app to ChatGPT and Claude using the tunnel.
The playground runs on Alpic: check the host compatibility matrix for which hooks it supports.

Go Further

DevTools

Call tools and render views locally, without a host

Tunnel

Expose your local server to real hosts

Audit

Catch spec and platform issues before submission

Connect to Hosts Audit

⌘I

​Start the Playground

​Exercise the Model Loop

​Limitations

​Go Further