MCP Server Cgb

🚀 MCP Server

The MCP Server is a server based on the Model Context Protocol (MCP), which allows Large Language Models (LLMs) to interact with external tools and services in a standard way.

🚀 Quick Start

The MCP Server enables seamless interaction between LLMs and external tools. You can quickly set up and run the server following the steps below.

✨ Features

• Support for multiple tools (calculator, weather, search)
• RESTful API based on FastAPI
• Data validation using Pydantic
• Efficient asynchronous processing
• Automatic generation of API documentation (with Swagger UI)

📦 Installation

System Requirements

• Python 3.8+
• pip (Python package manager)

Installation Steps

1. Clone the repository:

git clone <repository-url>
cd mcp-server

2. Create and activate a virtual environment:

python -m venv venv
# Windows
venv\Scripts\activate
# Linux/Mac
source venv/bin/activate

3. Install the required libraries:

pip install -r requirements.txt

💻 Usage Examples

Basic Usage

To run the server, execute the following command:

python main.py

The server will run at the following address: http://localhost:8000

API Endpoints

1. Root Endpoint

• URL: /
• Method: GET
• Description: Returns information about the server and a list of supported tools

2. Execute Tool

• URL: /execute
• Method: POST
• Description: Executes a tool with the provided parameters
• Request Body:

{
  "tool_name": "calculator",
  "parameters": {
    "operation": "add",
    "numbers": [1, 2, 3]
  },
  "context": {}
}

Supported Tools

1. Calculator

• Performs basic mathematical operations
• Parameters:
  • operation: "add" or "multiply"
  • numbers: An array of numbers to be calculated

2. Weather

• Retrieves weather information for a specific location
• Parameters:
  • location: The name of the location

3. Search

• Performs an internet search
• Parameters:
  • query: The search keyword

📚 Documentation

Access http://localhost:8000/docs to view the complete API documentation with Swagger UI.

🔧 Technical Details

Development

Adding a New Tool

1. Add the name and description of the new tool to SUPPORTED_TOOLS.
2. Add a new processing function in process_tool_request.
3. Create a separate processing function for the new tool.

Running Tests

pytest

📄 License

This project is licensed under the MIT License.