PixelFix

An MCP (Model Context Protocol) tool that gives AI the ability to read and analyze images.

Features

| Tool | Description | |------|-------------| | read_image | Read an image and return base64 content + metadata | | analyze_image | Analyze an image with a custom prompt | | image_metadata | Get detailed image metadata without loading full content | | list_images | Scan a directory for image files | | compare_images | Compare two images side by side |

Supports: PNG, JPG, JPEG, GIF, WebP, BMP, TIFF, SVG, ICO

Installation

Claude Code

claude mcp add -s user pixelfix -- npx pixelfix

Cursor / Windsurf / VS Code (Cline)

Add to your MCP config file:

{
  "mcpServers": {
    "pixelfix": {
      "command": "npx",
      "args": ["-y", "pixelfix"]
    }
  }
}

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "pixelfix": {
      "command": "npx",
      "args": ["-y", "pixelfix"]
    }
  }
}

Usage

Once configured, just ask your AI:

"Read the image at /path/to/image.png" "What's in this screenshot?" "Compare these two images" "List all images in /path/to/folder"

Development

npm install
npm run build
npm run inspector   # Test with MCP Inspector

License

MIT