Skip to main content
AI coding assistants work best when they have full context about the tools you are using. fal provides an MCP server that gives any compatible AI tool access to the complete fal documentation, so your assistant can generate code, debug issues, and discover features without you leaving your editor. Whether you are calling Model APIs or deploying your own models with Serverless, your AI assistant becomes an expert in all things fal. Model Context Protocol (MCP) is an open protocol that standardizes how applications provide context to LLMs. Any tool that supports MCP can connect to fal’s documentation server at https://docs.fal.ai/mcp. Today the setup guide below covers Cursor, but the same MCP server works with any compatible client.

​
Cursor

Add the fal MCP server to Cursor in four steps:
1

Open Command Palette

Use Cmd+Shift+P (Ctrl+Shift+P on Windows) to open the command palette.
2

Search for MCP Settings

Search for “Open MCP settings”.
3

Add Custom MCP

Select Add custom MCP. This will open the mcp.json file.
4

Configure fal Server

Add the following configuration, save the file, and restart Cursor:
{
  "mcpServers": {
    "fal": {
      "url": "https://docs.fal.ai/mcp"
    }
  }
}
Once connected, your AI assistant can answer questions about any fal model, API, or feature, generate code with full context from the documentation, debug faster with instant access to error explanations, and help you migrate from other platforms with contextual guidance.