FastMCP Echo Server & Client

This project demonstrates a complete FastMCP (Model Context Protocol) setup with both a server and client implementation.

FastMCP Echo Server

The server (fastmcp_echo_server.py) provides:

Tools

• echo_tool: Takes text input and returns it unchanged

Resources

• echo://static: Static resource that returns "Echo!"
• echo://{text}: Template resource that echoes the input text

Prompts

• echo: Takes text input and returns it as a prompt

MCP Client

The client (mcp_client.py) demonstrates how to:

• Initialize connection with MCP server
• List and call tools
• List and read resources
• List and get prompts
• Handle MCP protocol communication

Running the Demo

1. Start the FastMCP Server

# Install dependencies
pip install fastmcp

# Run the server
python fastmcp_echo_server.py

The server will start on http://localhost:8000 by default.

2. Run the MCP Client

In a separate terminal:

# Run the client (make sure server is running first)
python mcp_client.py

Expected Output

The client will demonstrate:

1. Server Connection: Initialize and connect to the MCP server
2. Tool Discovery: Find the echo_tool
3. Tool Execution: Call echo_tool with "Hello from MCP Client!"
4. Resource Discovery: Find both static and template resources
5. Resource Reading: Read echo://static and echo://hello
6. Prompt Discovery: Find the echo prompt
7. Prompt Execution: Get prompt with "Hello from MCP Client!"

Sample Output

🚀 Starting FastMCP Echo Server Demonstration
📡 Server URL: http://localhost:8000
🔄 Initializing connection with MCP server...
✅ Successfully connected to FastMCP server
   Server: Echo Server
   Version: 1.0.0

🛠️  Discovering tools...
✅ Found 1 tools
   • echo_tool: Echo the input text

🔧 Calling echo_tool...
✅ Tool call successful: Hello from MCP Client!

📁 Discovering resources...
✅ Found 2 resources
   • Static Echo Resource: echo://static
   • Template Echo Resource: echo://{text}

📖 Reading static resource: echo://static
✅ Resource read successful: Echo!

📖 Reading template resource: echo://hello
✅ Resource read successful: Echo: hello

💬 Discovering prompts...
✅ Found 1 prompts
   • echo: Echo prompt

💬 Getting echo prompt...
✅ Prompt get successful: Hello from MCP Client!

✅ Demonstration completed successfully!

MCP Protocol Details

The client uses the Model Context Protocol (MCP) with JSON-RPC 2.0 over HTTP to communicate with the server. It demonstrates:

• Initialization: Handshake and capability negotiation
• Tool Calling: Remote procedure calls to server tools
• Resource Access: Reading server-provided resources
• Prompt Generation: Getting structured prompts from the server

Customization

Server URL

To connect to a different server URL, modify BASE_URL in mcp_client.py:

BASE_URL = "http://your-server-url:port"

Server Implementation

The server uses the fastmcp library. Make sure it's installed:

pip install fastmcp

Requirements

• Python 3.8+
• fastmcp library
• requests library (for the client)

Install all dependencies:

pip install fastmcp requests