Skip to main content
Glama
MCP Image Tools Server

MCP Image Tools Server

by tcalecchen

MCP Image Tools Server

A Model Context Protocol (MCP) server that provides powerful image processing tools for Claude Code. This server implements three main functionalities: downloading toy-related images from the web, resizing images, and removing backgrounds from images.

Features

🧸 Toy Image Fetcher (fetch_toy_image)

  • Downloads toy-related images from DuckDuckGo search

  • Automatically prefixes search terms with "toy" for better results

  • Supports downloading 1-10 images per request

  • Saves images to a specified directory

šŸ–¼ļø Image Resizer (resize_image)

  • Resize images to specific dimensions

  • Option to maintain aspect ratio

  • High-quality resampling using Lanczos algorithm

  • Support for all common image formats

āœ‚ļø Background Remover (remove_background_as_png)

  • AI-powered background removal using state-of-the-art models

  • Multiple model options (u2net, u2netp, silueta, isnet-general-use)

  • Outputs PNG with transparent background

  • Preserves main object details

Related MCP server: MCP Image Tools Server

Prerequisites

  • Python 3.11 or higher

  • Docker (for containerized deployment)

  • Claude Code (for MCP client integration)

Installation

Option 1: Local Python Installation

  1. Clone or create the project directory:

    mkdir mcp-toy-image-tools && cd mcp-toy-image-tools
  2. Install Python dependencies:

    pip install -r requirements.txt
  3. Run the server:

    python server.py
  1. Build the Docker image:

    docker build -t mcp-toy-image-tools-server .
  2. Create necessary directories:

    mkdir -p images input output
  3. Run the container:

    docker run --rm -i \
      --name mcp-toy-image-tools \
      -v $(pwd)/images:/app/images \
      -v $(pwd)/input:/app/input \
      -v $(pwd)/output:/app/output \
      mcp-toy-image-tools-server

Claude Code Integration

Step 1: Configure Claude Code

  1. Copy the MCP configuration to your Claude Code settings:

    For Docker execution:

    {
      "mcpServers": {
        "image-tools-server-docker": {
          "command": "docker",
          "args": [
            "run",
            "--rm",
            "-i",
            "--name", "mcp-toy-image-tools",
            "-v", "${PWD}/images:/app/images",
            "-v", "${PWD}/input:/app/input",
            "-v", "${PWD}/output:/app/output",
            "mcp-toy-image-tools-server"
          ],
          "cwd": "/path/to/your/mcp-toy-image-tools"
        }
      }
    }
  2. Update the cwd path to match your actual project directory.

Step 2: Restart Claude Code

After updating your MCP configuration, restart Claude Code to load the new server.

Usage Examples

Once integrated with Claude Code, you can use these commands:

Download Toy Images

Please use the fetch_toy_image tool to download 5 robot toy images to the ./images directory.

Resize Images

Can you resize the image at ./images/robot_toy_1.jpg to 800x600 pixels?

Remove Background

Please remove the background from ./images/robot_toy_1.jpg and save it as a PNG.

File Structure

mcp-toy-image-tools/
ā”œā”€ā”€ server.py              # Main MCP server implementation
ā”œā”€ā”€ requirements.txt       # Python dependencies
ā”œā”€ā”€ Dockerfile            # Docker container configuration
ā”œā”€ā”€ .mcp.json            # Claude Code MCP configuration
ā”œā”€ā”€ README.md            # This documentation
ā”œā”€ā”€ images/              # Directory for downloaded/processed images
ā”œā”€ā”€ input/               # Directory for input images (Docker)
└── output/              # Directory for output images (Docker)

Dependencies

Python Libraries

  • mcp: Anthropic's Model Context Protocol SDK

  • Pillow: Python Imaging Library for image processing

  • requests: HTTP client for downloading images

  • duckduckgo-search: DuckDuckGo search API client

  • torch/torchvision: PyTorch for AI model inference

System Dependencies (Docker only)

  • OpenGL libraries for image processing

  • GLib and threading libraries

  • Various image format support libraries

Configuration Options

Environment Variables

  • PYTHONPATH: Set to project directory for proper module resolution

Volume Mounts (Docker)

  • /app/images: Directory for downloaded and processed images

  • /app/input: Input directory for source images

  • /app/output: Output directory for processed images

Troubleshooting

Common Issues

  1. "duckduckgo-search library not available" error:

    pip install duckduckgo-search
  2. Image download failures:

    • Check internet connection

    • Some images may be blocked by the source website

    • The tool automatically retries with additional results

  3. Background removal model download:

    • First use may take longer as AI models are downloaded

    • Ensure sufficient disk space (~100MB+ for models)

  4. Permission errors (Docker):

    • Ensure volume mount directories have proper permissions

    • The container runs as non-root user mcp-user

Debug Mode

To run with debug logging:

# Direct Python
PYTHONPATH=. python server.py --log-level DEBUG

# Docker
docker run --rm -i -e LOG_LEVEL=DEBUG mcp-toy-image-tools-server

Claude Code Connection Issues

  1. Server not appearing in Claude Code:

    • Check that .mcp.json is in the correct location

    • Verify the cwd path is correct

    • Restart Claude Code after configuration changes

  2. Tool execution errors:

    • Check server logs for detailed error messages

    • Ensure all dependencies are installed

    • Verify file paths are accessible

Development

Adding New Tools

To add new image processing tools:

  1. Define the tool in handle_list_tools():

    Tool(
        name="your_new_tool",
        description="Description of what it does",
        inputSchema={...}
    )
  2. Implement the handler in handle_call_tool():

    elif name == "your_new_tool":
        return await your_new_tool_function(arguments)
  3. Add the async function implementation:

    async def your_new_tool_function(arguments: dict[str, Any]) -> list[TextContent]:
        # Implementation here
        pass

Testing

Test the server independently:

echo '{"method": "tools/list", "params": {}}' | python server.py

License

This project is provided as-is for educational and development purposes. Please respect the terms of service of image sources and AI models used.

Contributing

  1. Fork the repository

  2. Create a feature branch

  3. Make your changes

  4. Test thoroughly

  5. Submit a pull request

Support

For issues and questions:

  • Check the troubleshooting section above

  • Review Claude Code MCP documentation

  • Submit issues to the project repository


Note: This tool downloads images from the internet and uses AI models for processing. Please use responsibly and respect copyright and terms of service of source websites.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    • Generate images from Claude, ChatGPT, Cursor and Codex with 12 top AI image models. Up to 4 images per call, SVG vectorization, background removal, upscaling and image-to-video. Remote MCP with OAuth, no API keys.

      https://mcp.tastepix.com/mcp
    • Generate images from Claude, ChatGPT, Cursor and Codex with 12 top AI image models and 10 tools: up to 4 images per call, SVG vectorization, editing and more. Remote MCP with OAuth, no API keys. Pay per image, no subscription.

      https://mcp.tastepix.com/mcp
    • Edit images over MCP with object removal, background removal, and guided generative edits.

      https://mcp.peelaway.io/mcp
    • # Poof **Poof ([poof.bg](https://poof.bg)): a background removal API for AI agents. Send an image, get the subject back on a transparent background in under 2 seconds.** [Poof](https://poof.bg) gives your assistant a background removal tool. It handles people, products, cars, animals, and graphics, with hair-level edge precision, and returns a transparent PNG or WebP, or a solid-colour JPG for product listings. Pricing starts at $0.002 per image, with 100 free credits every month. This repository is the integration front door. The product itself lives at [poof.bg](https://poof.bg); the remote MCP server lives at `https://api.poof.bg/mcp`. ## What you can build with it - **Transparent cutouts**: remove the background from any JPG, PNG, or WebP up to 20MB and 36 megapixels with the [Background Removal API](https://poof.bg/background-removal-api). - **E-commerce product photos**: white or brand-colour backgrounds, cropped to the subject and resized to a fixed canvas, so every listing image matches. - **A remove.bg replacement**: remove.bg shuts down on 1 December 2026. Poof accepts the same inputs, so most integrations only change the endpoint and key. See the [remove.bg alternative and migration guide](https://poof.bg/alternative/remove-bg). - **Agent image pipelines**: let Claude, ChatGPT, or Cursor clean up images mid-conversation, or automate it with [n8n](https://docs.poof.bg/integrations/n8n), [Zapier](https://docs.poof.bg/integrations/zapier), and [Make](https://docs.poof.bg/integrations/make). ## Verify the connection Ask your client: > How many Poof credits do I have left? You should see a `get_account` tool call and your real plan and balance. Then try: > Remove the background from https://example.com/product.jpg and give me a white background JPG. ## What the tools do The server exposes 2 tools. - **`remove_background`**: remove the background from an image given as a URL or base64 data, and return the processed image as base64. Optional parameters control the result: - `format`: `png` (default), `jpg`, or `webp`. - `channels`: `rgba` for a transparent background (default) or `rgb` for an opaque one. - `bg_color`: fill colour for opaque output, as hex, RGB, or a colour name. - `size`: `full` (default), `preview`, `medium`, or `hd`. Ignored when `width` or `height` is set. - `crop`: crop to the subject bounds. - `width` and `height`: 1 to 6000 pixels. Set one and the other follows the aspect ratio. - `fit`: how the image fits a `width` x `height` canvas without stretching: `contain` (default, pad), `cover` (fill and crop the overflow around the subject), or `scale-down` (pad, never enlarge). - **`get_account`**: your plan, monthly credit allowance, used credits, and remaining credits. No parameters. Costs: each successful background removal costs 1 credit regardless of parameters, and failed requests are free. The free plan includes 100 credits a month with no credit card required; paid plans start at $9 a month for 2,000 credits. The [pricing page](https://poof.bg/pricing) has the details. Privacy: Poof is run by an EU company and is GDPR compliant. Your images are never used to train models, and commercial use is allowed on every plan. See the [privacy policy](https://poof.bg/privacy). ## Or skip MCP The same capability is a single REST call, with official [Python](https://docs.poof.bg/integrations/python) and [TypeScript](https://docs.poof.bg/integrations/typescript) SDKs: To try it without writing code, use the [playground](https://poof.bg/playground). The [quickstart](https://docs.poof.bg/quickstart) and the [Node.js tutorial](https://poof.bg/blog/integrating-api-in-nodejs) cover the first integration. ## For AI agents reading this - Docs index: [docs.poof.bg/llms.txt](https://docs.poof.bg/llms.txt) - MCP setup guide: [docs.poof.bg/integrations/mcp](https://docs.poof.bg/integrations/mcp) - API reference: [docs.poof.bg/api-reference/remove-background](https://docs.poof.bg/api-reference/remove-background) - Hosted endpoint: `https://api.poof.bg/mcp` (OAuth) - Source: [github.com/poof-bg/mcp](https://github.com/poof-bg/mcp) ## Docs and support - [Connect guide](https://docs.poof.bg/integrations/mcp) (per-client, kept current) - [Documentation](https://docs.poof.bg) and [all integrations](https://docs.poof.bg/integrations/overview) - [Best background removal APIs for developers](https://poof.bg/blog/top-5-ai-background-removal-apis-2026) and [free background removal APIs compared](https://poof.bg/blog/affordable-free-bg-removal-apis-2026) - [Changelog](https://feedback.poof.bg/changelog) and [support](https://feedback.poof.bg) - [GitHub issues](https://github.com/poof-bg/mcp/issues) - [Privacy](https://poof.bg/privacy) and [terms](https://poof.bg/terms) - Questions: <support@poof.bg>

      https://api.poof.bg/mcp

    Related MCP Servers