Playground

The Playground is a full-screen chat interface for trying SpiderGate interactively — pick a model, send messages, watch the response stream, then copy the equivalent API code. It's the fastest way to test a prompt or compare models before wiring anything up. Open it at /dashboard/gate/playground.

What you can do

• Choose a model. Select any task alias or concrete model from the picker and see how each responds to the same prompt.

• Stream responses. Replies stream token-by-token, exactly as the streaming API delivers them — so what you see is what your app will get.

• Get the code. A Get Code panel shows the exact request for what you just ran, ready to paste into curl or the OpenAI SDK. Tune the prompt in the UI, then lift the working call straight into your code.

A typical loop

1. Open the Playground and pick a model — start with spideriq/fast for quick iteration.

2. Write a system prompt and a user message; send it.

3. Not happy with the result? Switch the model (try spideriq/coding or agent/chat) and resend the same prompt to compare.

4. Once it's right, open Get Code and copy the request into your application.

When to use it

The Playground is for exploration — finding the right model and prompt. For production traffic, call the API directly with an agent key, and review what ran in Traces.

Next steps

1. Found the right prompt? Copy it into a real request — Chat Completions.

2. Compare models with the Task Aliases reference.

3. Review your test requests in Traces.