Skip to main content

​
Overview

Framelane exposes an MCP (Model Context Protocol) server at: 
https://api.framelane.io/mcp
Any MCP host — Claude Desktop, Cursor, LangChain, CrewAI, etc. — can connect to this endpoint and use Framelane tools directly without writing API integration code.

​
Available tools

ToolDescription
create_renderSubmit a render job
get_renderPoll render status and progress
create_taskSubmit any AI processing task
get_taskPoll task status
list_artifactsGet signed download URLs
register_webhookSubscribe to job events

​
Authentication

Pass your API key as a Bearer token in the MCP connection headers: 
{
  "mcpServers": {
    "framelane": {
      "url": "https://api.framelane.io/mcp",
      "headers": {
        "Authorization": "Bearer fl_live_..."
      }
    }
  }
}

​
Connecting in Cursor

Add to your Cursor MCP settings (~/.cursor/mcp.json): 
{
  "mcpServers": {
    "framelane": {
      "url": "https://api.framelane.io/mcp",
      "headers": {
        "Authorization": "Bearer ${FRAMELANE_API_KEY}"
      }
    }
  }
}
Then ask Cursor: “Use the Framelane MCP to render this composition…”

​
Connecting in Claude Desktop

Add to claude_desktop_config.json: 
{
  "mcpServers": {
    "framelane": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://api.framelane.io/mcp"],
      "env": {
        "FRAMELANE_API_KEY": "fl_live_..."
      }
    }
  }
}

​
Agent Skills via MCP

The Framelane Agent Skills are exposed as MCP resources — loadable playbooks that encode the full workflow for common operations. Load them from: 
https://docs.framelane.io/mcp