MCP Servers Template

This repository provides a template for setting up Model Context Protocol (MCP) servers for local development. It includes examples of different MCP server implementations:

1. Weather API - Provides weather alerts and forecasts
2. Calculator - Basic and advanced mathematical operations
3. Notes - Simple note-taking functionality with local file storage

Note: This project requires Python 3.10 or higher. See footnotes for instructions on using the correct Python version.

Project Structure

├── servers/          # MCP server implementations
│   ├── calculator.py # Calculator server implementation
│   ├── notes.py      # Notes server implementation
│   └── weather.py    # Weather API server implementation
├── docs/             # Documentation for the servers
├── requirements.txt  # Python dependencies
└── README.md         # This file

Requirements

• Python 3.10+
• FastMCP library
• Other dependencies as listed in requirements.txt
• UV package manager (recommended for faster installations)

Setup

1. Clone the repository:

   git clone <repository-url>
   cd mcp-server-templates
   

2. Set up a virtual environment (recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   

3. Install UV package manager (if not already installed):

   pip install uv
   

4. Install dependencies with UV:

   uv pip install -r requirements.txt
   

Usage

Running the MCP Servers

Each MCP server can be run using the MCP dev server command:

Weather API Server

mcp dev servers/weather.py

The Weather API provides the following tools:

• get_alerts: Get weather alerts for a US state
• get_forecast: Get weather forecast for a location based on latitude and longitude

Calculator Server

mcp dev servers/calculator.py

The Calculator provides mathematical operations including:

• Basic operations: add, subtract, multiply, divide
• Advanced operations: power, sqrt, cbrt, factorial, log, etc.
• Trigonometric functions: sin, cos, tan

Notes Server

mcp dev servers/notes.py

The Notes server provides note management functionality including:

• add_note: Create a new note with title and content
• list_notes: Get all saved notes
• get_note: Retrieve a specific note by ID
• delete_note: Remove a note by ID
• search_notes: Find notes containing a specific keyword
• update_note: Modify an existing note's title or content

VS Code Integration

These MCP servers can be integrated with Visual Studio Code for use with AI assistants. Configure your settings.json to include:

{
  "mcp": {
    "servers": {
      "weather": {
        "command": "bash",
        "args": [
          "-c",
          "source {AbsolutePathToTheFolder}/venv/bin/activate && uv --directory {AbsolutePathToTheFolder} run servers/weather.py"
        ]
      },
      "calculator": {
        "command": "bash",
        "args": [
          "-c",
          "source {AbsolutePathToTheFolder}/venv/bin/activate && uv --directory {AbsolutePathToTheFolder} run servers/calculator.py"
        ]
      },
      "notes": {
        "command": "bash",
        "args": [
          "-c",
          "source {AbsolutePathToTheFolder}/venv/bin/activate && uv --directory {AbsolutePathToTheFolder} run servers/notes.py"
        ]
      }
    }
  }
}

License

This project is licensed under the Apache License 2.0. See the LICENSE file for details.

Footnotes

Using Python 3.10+

1. Check your Python version:

   python --version
   

2. Install Python 3.10 or higher if needed:

   • macOS (using Homebrew):

     brew install python@3.10
     

   • Linux:

     sudo apt-get update
     sudo apt-get install python3.10
     

   • Windows: Download installer from python.org

3. Use pyenv for managing multiple Python versions (recommended):

   # Install pyenv
   brew install pyenv  # macOS
   
   # Install Python version
   pyenv install 3.10.0
   
   # Set local Python version
   pyenv local 3.10.0
   

4. Using UV with pyenv:

   # After setting up pyenv and activating your environment
   pip install uv
   uv pip install -r requirements.txt
   

References and Further Reading

• MCP Introduction & Official Documentation
• MCP Python Package on PyPI