MCP Tool Server Template
Description
# MCP Tool Server Template A cookiecutter template for creating Model Context Protocol (MCP) tool servers with FastAPI, uv, and Docker support. ## Features - FastAPI-based REST server - Model Context Protocol (MCP) tool support - Sample weather tool implementation - Modern…
About
# MCP Tool Server Template A cookiecutter template for creating Model Context Protocol (MCP) tool servers with FastAPI, uv, and Docker support. ## Features - FastAPI-based REST server - Model Context Protocol (MCP) tool support - Sample weather tool implementation - Modern Python tooling (uv, ruff, mypy) - Type hints…
Details
- Author
- mibexx
- Downloads
- 179
- Categories
- Developer Tools
Jump to
- FastAPI-based REST server with MCP tool support
- Sample weather tool for quick prototyping
- Modern Python tooling: uv, ruff, mypy
- Type hints and comprehensive testing
- Docker and Docker Compose support
- Kubernetes deployment configuration
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
MCP Tool Server TemplateCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Use cookiecutter to generate a new project: cookiecutter gh:mibexx/mbxai-mcp-server-template and answer the prompts for project name, slug, etc. Then navigate into the generated directory and run uv run service to start the server locally, or use docker-compose up -d for containerized execution.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"mcp tool server template": {
"mbxai-mcp-server-template": {
"command": "uv",
"args": [
"venv"
]
}
}
}
}McpServers
{
"mbxai-mcp-server-template": {
"command": "uv",
"args": [
"venv"
]
}
}MCP Tool Server Template
A cookiecutter template for creating Model Context Protocol (MCP) tool servers with FastAPI, uv, and Docker support.
Features
- FastAPI-based REST server
- Model Context Protocol (MCP) tool support
- Sample weather tool implementation
- Modern Python tooling (uv, ruff, mypy)
- Type hints and comprehensive testing
- Docker support for containerization
- Kubernetes deployment configuration
Requirements
- Python 3.12 or higher
- uv package manager
- cookiecutter 2.5.0 or higher
Usage
Creating a New Project
cookiecutter gh:mibexx/mbxai-mcp-server-template
Follow the prompts to configure your project:
- project_name: The name of your project
- project_slug: The slug for your project (used in paths and imports)
- package_name: The name of your Python package
- description: A short description of your project
- author_name: Your name
- author_email: Your email
- open_source_license: The license for your project
- python_version: The Python version to use
- use_pytest: Whether to include pytest for testing
- use_mypy: Whether to include mypy for type checking
- use_ruff: Whether to include ruff for linting
Project Structure
The template creates a well-organized project structure:
your-project/
├── src/
│ └── your_package/
│ ├── api/ # Core API functionality
│ ├── clients/ # Client libraries
│ ├── project/ # Project-specific code
│ │ └── weather.py # Sample weather tool
│ ├── config.py # Configuration management
│ └── __init__.py # Package initialization
├── tests/ # Test suite
├── data/ # Data storage
├── logs/ # Log files
├── kubernetes/ # Kubernetes deployment files
├── Dockerfile # Docker configuration
├── docker-compose.yml # Docker Compose configuration
├── pyproject.toml # Project metadata and dependencies
└── README.md # Project documentation
Development
Running Locally
```bash
cd your-project
uv venv
source .venv/bin/activate # On Unix/macOS
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
