Fifth Elephant MCP Server

Fifth Elephant MCP Server: A DIY Guide

This guide provides a complete walkthrough to create and run a simple Model Context Protocol (MCP) server from scratch.

Prerequisites

Before we begin, ensure you have the following installed:

1. Install pyenv (if not already installed)

pyenv allows you to manage multiple Python versions easily.

macOS:

brew install pyenv

Linux:

curl https://pyenv.run | bash

Windows:

# Install pyenv-win using PowerShell
Invoke-WebRequest -UseBasicParsing -Uri "https://raw.githubusercontent.com/pyenv-win/pyenv-win/master/pyenv-win/install-pyenv-win.ps1" -OutFile "./install-pyenv-win.ps1"; &"./install-pyenv-win.ps1"

Add pyenv to your shell profile:

macOS/Linux (zsh):

echo 'export PYENV_ROOT="$HOME/.pyenv"' >> ~/.zshrc
echo 'command -v pyenv >/dev/null || export PATH="$PYENV_ROOT/bin:$PATH"' >> ~/.zshrc
echo 'eval "$(pyenv init -)"' >> ~/.zshrc
source ~/.zshrc

Windows:

# Restart your PowerShell or Command Prompt after installation
# pyenv-win will be automatically added to your PATH

2. Install Python 3.11

Install and set Python 3.11 as your local version:

pyenv install 3.11.9
pyenv local 3.11.9

Verify the Python version:

python --version
# Should output: Python 3.11.9

3. Install uv

Install uv for fast Python package management:

macOS/Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows:

powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

Verify uv installation:

uv --version

Step 1: Project Setup

First, create a directory for your project and navigate into it.

mkdir fifth-elephant-mcp
cd fifth-elephant-mcp

Step 2: Initialize a Python Project with `uv`

Initialize a new Python project using uv. This will create a virtual environment and a pyproject.toml file.

uv init --quiet

Step 3: Create and Activate Virtual Environment

Create a virtual environment using Python 3.11:

uv venv --python 3.11.9

Activate the virtual environment:

macOS/Linux:

source .venv/bin/activate

Windows (Command Prompt):

.venv\Scripts\activate.bat

Verify you're in the virtual environment (you should see (.venv) in your terminal prompt).

Step 4: Sync Dependencies (not required)

Sync the project dependencies to ensure everything is up to date:

uv sync

Step 5: Add MCP Dependency

Add the mcp library with CLI extras to your project's dependencies.

uv add "mcp[cli]"

Note: When installing MCP CLI, it may ask you to install additional dependencies. Press y to confirm.

Step 6: Verify Installation

Verify that all components are properly installed:

# Verify Python version
python --version
# Should output: Python 3.11.9

# Verify uv version
uv --version

# Verify MCP library is installed (should be version 1.11 or later)
python -c "from mcp.server.fastmcp import __version__; print(__version__)"    

Step 7: Edit the Server Code

When you ran uv init, a main.py file was already created. Edit this file and replace its contents with the following code. This code sets up a FastMCP server with two tools: hello_world and add.

from mcp.server.fastmcp import FastMCP

# Create an MCP server
mcp = FastMCP("Fifth Elephant")

# Add a hello world tool
@mcp.tool()
def hello_world() -> str:
    """Returns a friendly greeting."""
    return "hello world"

# Add an addition tool
@mcp.tool()
def add(a: int, b: int) -> int:
    """Adds two numbers together."""
    return a + b

if __name__ == "__main__":
    mcp.run()

Step 8: Debug Locally with MCP Inspector

For development and testing, you can use the MCP inspector.

Run the following command -

uv run mcp dev main.py

This will spin up the inspector. It will open in your browser. Press connect on the left sidebar.

Step 9: Configure Claude Desktop

If you haven't already, install Claud