Liveblocks icon

Liveblocks

Call the Liveblocks REST API (rooms, comments, feeds, Yjs, AI, and more)

AI Copilot → Create

AI-generated

Summary

Create a new AI Copilot configuration on the Liveblocks platform, specifying provider details, prompts, and API key to configure its behavior.

Inputs

  • Provider (required) — Select which AI provider to use (OpenAI, Anthropic, Google, or OpenAI Compatible).
  • Name (required) — The name identifier for the AI Copilot instance.
  • Description — Optional descriptive text about this AI Copilot.
  • System Prompt (required) — The system prompt to guide the AI Copilot's behavior.
  • Knowledge Prompt — Optional knowledge prompt to provide contextual information to the AI Copilot.
  • Always Use Knowledge — Whether to always include the knowledge prompt in interactions.
  • Provider API Key (required) — API key to authenticate requests to the selected AI provider.
  • Provider Settings — JSON object of provider-specific settings such as max tokens and temperature.
  • Provider Model (required) — The model ID to use for the AI provider (e.g., gpt-4o, claude-sonnet-4-5-20250929, gemini-2.5-flash).
  • Compatible Provider Name — If the provider is OpenAI Compatible, specify the compatible provider name.
  • Provider Base URL — Base URL for the API of an OpenAI Compatible provider, if applicable.
  • Provider Options (JSON) — Additional provider-specific options stored as JSON.
  • Include Request in Error — Flag to include detailed request info in error responses for debugging.

Output shape

a single record representing the created AI Copilot configuration returned from the Liveblocks API

The response includes the configured AI Copilot details as confirmed by the remote service. Error details may include request data if 'Include Request in Error' is enabled.

Links

Discussion