Skip to main content
The fastest way to get started is connecting to the hosted server. No installation required. Below are setup instructions for each supported AI tool.

​
Claude Desktop

Add to your Claude Desktop config:
  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "lifi": {
      "type": "http",
      "url": "https://mcp.li.quest/mcp"
    }
  }
}
With an API key for higher rate limits: 
{
  "mcpServers": {
    "lifi": {
      "type": "http",
      "url": "https://mcp.li.quest/mcp",
      "headers": {
        "X-LiFi-Api-Key": "your_api_key"
      }
    }
  }
}

​
Claude Code

Add via CLI: 
claude mcp add lifi --transport http --url https://mcp.li.quest/mcp
Or add a .mcp.json file at your project root (shareable via git): 
{
  "servers": {
    "lifi": {
      "type": "http",
      "url": "https://mcp.li.quest/mcp"
    }
  }
}

​
Cursor

Add to .cursor/mcp.json (project-level) or ~/.cursor/mcp.json (global): 
{
  "mcpServers": {
    "lifi": {
      "url": "https://mcp.li.quest/mcp"
    }
  }
}

​
Windsurf

Add to ~/.codeium/windsurf/mcp_config.json: 
{
  "mcpServers": {
    "lifi": {
      "serverUrl": "https://mcp.li.quest/mcp"
    }
  }
}
Windsurf uses serverUrl instead of url.

​
VS Code (GitHub Copilot)

Add to .vscode/mcp.json in your project: 
{
  "servers": {
    "lifi": {
      "type": "http",
      "url": "https://mcp.li.quest/mcp"
    }
  }
}
With secure API key prompting: 
{
  "inputs": [
    {
      "id": "lifi-api-key",
      "type": "promptString",
      "description": "LI.FI API Key",
      "password": true
    }
  ],
  "servers": {
    "lifi": {
      "type": "http",
      "url": "https://mcp.li.quest/mcp",
      "headers": {
        "X-LiFi-Api-Key": "${input:lifi-api-key}"
      }
    }
  }
}

​
ChatGPT

ChatGPT supports MCP via developer mode. Add the server through the ChatGPT UI. There is no config file to edit.
  1. Open ChatGPT settings
  2. Navigate to the MCP servers section
  3. Add the server URL: https://mcp.li.quest/mcp

​
Stdio Transport (Self-Hosted / Local)

For local MCP clients running the binary directly: 
go install github.com/lifinance/lifi-mcp@latest
Add to your MCP client config: 
{
  "mcpServers": {
    "lifi": {
      "command": "lifi-mcp",
      "env": {
        "LIFI_API_KEY": "your_api_key"
      }
    }
  }
}
In stdio mode, the API key is read from the LIFI_API_KEY environment variable. Without it, the server uses the public rate limit (200 requests / 2 hours).

​
Self-Hosted HTTP Transport

If you’re running your own instance of the MCP server: 
{
  "mcpServers": {
    "lifi": {
      "url": "http://localhost:8080/mcp"
    }
  }
}
With an API key: 
{
  "mcpServers": {
    "lifi": {
      "url": "http://localhost:8080/mcp",
      "headers": {
        "X-LiFi-Api-Key": "your_api_key"
      }
    }
  }
}
Start the server locally with: 
go install github.com/lifinance/lifi-mcp@latest
lifi-mcp --transport http --port 8080
Or with Docker: 
docker run -p 8080:8080 ghcr.io/lifinance/lifi-mcp

GitHub Repository

Full source code, Docker setup, and self-hosting documentation

​
Testing Your Setup

Use the MCP Inspector to interactively test the server: 
npx @modelcontextprotocol/inspector --url https://mcp.li.quest/mcp
This opens a web UI at http://localhost:6274 where you can browse and test all available tools. You can also verify your setup by asking your AI tool a simple question like:
“What chains does LI.FI support?”
If the MCP server is connected correctly, the tool will call get-chains and return a list of supported blockchains.