# Quiver AI MCP Server

An MCP server that exposes [Quiver AI](https://quiver.ai) vector graphics APIs as tools, built with [XMCP](https://xmcp.dev) and deployable to Railway.

## Tools

| Tool | Description |
|------|-------------|
| `generate_svg` | Generate SVG vector graphics from a text prompt, with optional reference images |
| `vectorize_image` | Convert a raster image (PNG, JPG) into an SVG vector graphic |
| `list_models` | List available Quiver AI models |

## Setup

1. Copy `.env.example` to `.env` and fill in your keys:
   - `QUIVER_API_KEY` — from [Quiver AI](https://quiver.ai)
   - `MCP_API_KEY` — a strong random string you generate (used to authenticate MCP clients)

2. Install and run:
   ```bash
   npm install
   npm run dev
   ```

3. The MCP endpoint is available at `http://localhost:3000/mcp`

## Deploy to Railway

1. Push this repo to GitHub
2. Create a new Railway project from the repo
3. Set environment variables in Railway dashboard:
   - `QUIVER_API_KEY`
   - `MCP_API_KEY`
4. Railway will build and deploy using the Dockerfile

## Connect from Claude Desktop

Add to your Claude Desktop MCP config:

```json
{
  "mcpServers": {
    "quiver-ai": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://your-railway-domain.up.railway.app/mcp",
        "--header",
        "Authorization:Bearer YOUR_MCP_API_KEY"
      ]
    }
  }
}
```