Playground

The Playground is your private sandbox to experiment with your AI Agent in real time. It allows you to simulate conversations and verify that the Agent behaves exactly as you intended before it goes live on your website.

Whenever you make changes to your Agent’s settings—like switching the AI model, updating the System Prompt, uploading new knowledge base documents, or enabling new Actions—you can instantly test those changes in the Playground.

How to use the Playground

1. Open your workspace and navigate to the Agents page.
2. Select your Agent and click the Playground tab from the top navigation menu.
3. Start chatting with the Agent just like a normal user would.
4. Ask it questions that rely on your Knowledge Base to ensure it retrieves the correct answers.
5. Trigger scenarios that require your configured Actions (e.g., asking it to “book a meeting” or “contact support”) to verify the UI widgets and webhooks fire correctly.

If the Agent’s responses aren’t what you expect, you can simply switch back to the Settings or Knowledge Base tabs, tweak your instructions, and return to the Playground to instantly see the updated behavior.