MCP Tools Server

# MCP Tools Server

A simple MCP (Model Context Protocol) server built with Python and FastMCP.

This project is beginner-friendly and demonstrates how to create MCP tools, run an MCP server over HTTP, and test the tools using MCP Inspector.

✨ Features

This server currently provides the following tools:

1. Add Number

Tool: Add_number

Adds two numbers and returns their sum.

Example:

a = 10
b = 20
Result = 30

2. Random Number

Tool: random_no

Generates a random integer between the provided minimum and maximum values.

Default values:

min_val = 1
max_val = 100

Example:

min_val = 1
max_val = 50
Result = 37

3. Todo Task Management

The server also includes a simple in-memory Todo system.

Available tools:

Example task:

{
  "id": 1,
  "title": "Learn MCP",
  "completed": false
}

Note: Todo data is stored in memory using a Python list. Data will be lost when the server is restarted.

Related MCP server: MCP Server Basic Example

📦 MCP Resource

The server also exposes:

info://server

This resource returns basic server information such as server name, environment, host, port, operating system, runtime, and version.

✅ Requirements

• Python 3.10+

• uv

• FastMCP

• Node.js and npx (for MCP Inspector)

🚀 Installation

Clone the repository:

git clone <YOUR_GITHUB_REPOSITORY_URL>
cd <YOUR_PROJECT_FOLDER>

Create and use the virtual environment with uv:

uv sync

If dependencies are not already defined, install FastMCP:

uv add fastmcp

▶️ Run the MCP Server

If your Python file is named main.py:

uv run main.py

The server runs using Streamable HTTP on:

http://127.0.0.1:8000/mcp

If you are using the Todo server example on port 8001, use:

http://127.0.0.1:8001/mcp

🔍 Test with MCP Inspector

Start MCP Inspector in another terminal:

npx -y @modelcontextprotocol/inspector

Open the Inspector URL shown in the terminal, usually:

http://localhost:6274

Add the server manually with:

Then connect to the server.

Go to the Tools tab to test:

• Add_number

• random_no

• add_task

• list_task

• completed_task

• delete_task

Go to the Resources tab to test:

• info://server

📸 MCP Inspector

Example of the server connected to MCP Inspector.

📁 Project Structure

A simple project can look like this:

MCP Server/
│
├── main.py
├── pyproject.toml
├── uv.lock
├── README.md
└── mcp-inspector.png

⚙️ How It Works

The basic flow is:

MCP Inspector
      |
      | Streamable HTTP
      v
FastMCP Server
      |
      +---- Add_number
      |
      +---- random_no
      |
      +---- add_task
      |
      +---- list_task
      |
      +---- completed_task
      |
      +---- delete_task
      |
      +---- info://server

🎯 Learning Goals

This project is useful for learning:

• What an MCP server is

• How to create MCP tools with @mcp.tool

• How tool arguments work

• How to expose resources with @mcp.resource

• How to run an MCP server using HTTP

• How to connect an MCP server to MCP Inspector

• How multiple tools can be exposed from one MCP server

• Basic CRUD-style application logic with an in-memory data store

👤 Author

Built as a learning project while exploring MCP and FastMCP.