🎨 Leonardo AI MCP Server

A remote Model Context Protocol (MCP) server for Leonardo.ai, powered by Vercel.

Generate images, manage models, check generation status, upscale, create variations — all through MCP from Cursor, Claude Desktop, VS Code, or any MCP-compatible client.

✨ Features / Tools

Related MCP server: imagengen

🚀 Deploy to Vercel

One-Click Deploy

Manual Deploy

1. Fork/clone this repository

2. Import it in Vercel

3. Set environment variable LEONARDO_API_KEY with your Leonardo AI API key

4. Enable Fluid Compute in your Vercel project settings for optimal performance

5. Deploy — your MCP server will be live at https://your-project.vercel.app/mcp

🔑 Get a Leonardo AI API Key

1. Go to Leonardo.ai

2. Sign up / log in

3. Navigate to API Access in your settings

4. Generate an API key

🔌 Connect Your MCP Client

Cursor

Add to your Cursor MCP settings (.cursor/mcp.json):

{
  "mcpServers": {
    "leonardo-ai": {
      "url": "https://your-project.vercel.app/mcp",
      "headers": {
        "x-leonardo-api-key": "YOUR_LEONARDO_API_KEY"
      }
    }
  }
}

Claude Desktop

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "leonardo-ai": {
      "url": "https://your-project.vercel.app/mcp",
      "headers": {
        "x-leonardo-api-key": "YOUR_LEONARDO_API_KEY"
      }
    }
  }
}

VS Code (GitHub Copilot)

Add to your VS Code MCP settings:

{
  "mcp": {
    "servers": {
      "leonardo-ai": {
        "url": "https://your-project.vercel.app/mcp",
        "headers": {
          "x-leonardo-api-key": "YOUR_LEONARDO_API_KEY"
        }
      }
    }
  }
}

Note: You can also set LEONARDO_API_KEY as an environment variable on Vercel instead of passing it as a header. If the env var is set, the header is optional.

🛠 Local Development

# Install dependencies
pnpm install

# Set your API key
cp .env.example .env
# Edit .env and add your LEONARDO_API_KEY

# Start local dev server
pnpm dev

The MCP server will be available at http://localhost:3000/mcp.

📁 Project Structure

├── api/
│   ├── server.ts           # MCP server with all tool definitions
│   └── leonardo-client.ts  # Leonardo AI REST API client
├── public/
│   └── index.html          # Landing page
├── package.json
├── tsconfig.json
├── vercel.json             # Vercel deployment config
├── .env.example
└── README.md

📋 API Key Options

The server resolves the Leonardo API key in this order:

1. x-leonardo-api-key header — passed by the MCP client (recommended for multi-user setups)

2. LEONARDO_API_KEY env variable — set in Vercel project settings (simpler for single-user)

🧪 Example Usage

Once connected, you can use natural language in your MCP client:

• "Generate an image of a sunset over mountains in cinematic style"

• "List available Leonardo AI models"

• "Show me my recent generations"

• "Upscale the last generated image"

• "Create a variation of image xyz"

📝 License

MIT