remote-mcp-server by CampbellJohn - MCP Server

Remote MCP Server

A simple client-server application using the MCP (Model Context Protocol) framework with Docker containerization.

Project Overview

This project demonstrates a simple MCP server with client connectivity. The server exposes tools that can be called remotely by the client. The entire application is containerized using Docker for easy deployment and consistent environments.

Features

MCP Server: Exposes API endpoints as tools that can be called remotely
MCP Client: Connects to the server and calls the available tools
Modular Tool Architecture: Server tools are organized in separate files for better maintainability
Docker Support: Both client and server are containerized
Docker Compose: Easy orchestration of the client and server containers

Example Tools

Hello World

A simple greeting tool that returns a hello message.

Parameters:

name (string): The name to greet

Returns:

A greeting message string

Calculator

Performs basic arithmetic operations.

Parameters:

operation (string): The operation to perform (add, subtract, multiply, divide)
a (float): First number
b (float): Second number

Returns:

A dictionary containing the operation and result

Getting Started

Prerequisites

Docker and Docker Compose
Python 3.12 (if running locally)

Running with Docker

Clone the repository:

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

Build and run the containers:
```
docker-compose up --build
```
The server will be available at http://localhost:8000/mcp

Running Locally

Server

Navigate to the server directory:
```
cd server
```
Install dependencies:
```
pip install -r requirements.txt
```
Run the server:
```
python server.py
```

Client

Navigate to the client directory:
```
cd client
```
Install dependencies:
```
pip install -r requirements.txt
```
Run the client:
```
python client.py
```

Adding New Tools

To add a new tool to the server:

Create a new Python file in the server/tools directory
Define a register_tool function that takes a server parameter
Use the server decorator to define your tool
Import and register your tool in server/tools/__init__.py

Example:

# server/tools/new_tool.py
from mcp.server.fastmcp import Context

def register_tool(server):
    @server.tool(
        name="new_tool",
        description="Description of the new tool"
    )
    def new_tool(param1: str, param2: int, ctx: Context) -> dict:
        # Tool implementation
        return {"result": "some result"}

Then update __init__.py:

from . import new_tool

def register_all_tools(server):
    # ... existing tools
    new_tool.register_tool(server)

CampbellJohn/remote-mcp-server

Hello World

Calculator