Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Toolkits MCPremove the background from product-image.png"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Toolkits MCP
A stdio MCP implementation using Fast MCP 3 meant to be run locally. This MCP provides a collection of useful tools for image processing.
MCP Tools
Image optimizer: Optimize images by reducing file size while maintaining quality.
Remove background from image: Remove the background from images using AI.
Favicon generator: Generate favicons of sizes 16, 32, 192, 512 from a logo.png file.
Tech Stack
uvfor Python environment managementPython
Fast MCP 3.0
PIL (Pillow) for image processing
rembg for background removal
Installation
Ensure you have
uvinstalled. If not, install it from https://github.com/astral-sh/uv.Clone or navigate to the project directory.
Install dependencies:
uv sync
Running Locally
To run the MCP server locally for testing:
This starts the server using stdio transport, which is suitable for local development.
Using in VSCode
To use this MCP server in VSCode, you can integrate it with the Claude Code extension or GitHub Copilot Chat that supports MCP.
Option 1: Using Claude Code Extension
Install the "Claude Code" extension in VSCode from the marketplace.
Create a
.mcp.jsonconfiguration file in your project root or user directory (e.g.,~/.mcp.jsonfor user scope).Add the following configuration to
.mcp.json:{ "mcpServers": { "toolkits-mcp": { "command": "uv", "args": ["run", "python", "main.py"], "cwd": "/path/to/toolkits-mcp" } } }Replace
/path/to/toolkits-mcpwith the actual path to your project directory.Restart VSCode or reload the Claude Code extension.
The tools should now be available in Claude Code conversations.
Option 2: Using GitHub Copilot Chat
Ensure you have GitHub Copilot Chat installed in VSCode.
Open VSCode settings (Cmd/Ctrl + ,) and search for "mcp".
In the "GitHub Copilot Chat: Mcp" setting, add the server configuration:
{ "toolkits-mcp": { "command": "uv", "args": ["run", "python", "main.py"], "cwd": "/path/to/toolkits-mcp" } }The tools will be available in Copilot Chat.
Usage
Once configured, you can use the tools in your AI assistant conversations:
To optimize an image: Ask the assistant to optimize an image file.
To remove background: Request background removal from an image.
To generate favicons: Provide a logo.png and request favicon generation.
Development
To contribute or modify the tools:
Make changes to
main.py.Test locally using
uv run python main.py.Ensure all dependencies are listed in
pyproject.toml.
License
[Add license information if applicable]