MCP Demo Project

A collection of Model Context Protocol (MCP) servers demonstrating different capabilities including math operations, weather data, and LangGraph workflows.

🚀 Quick Start

Prerequisites

• Python 3.8+

• Node.js (for MCP Inspector)

• Virtual environment activated

Installation

1. Clone and setup:

   git clone https://github.com/vaibhavGala262/MCP_servers.git 
   cd mcp_demo2

2. Create and activate virtual environment:

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

3. Install dependencies:

   pip install -r requirements.txt

4. Install MCP Inspector (globally):

   npm install -g @modelcontextprotocol/inspector

Related MCP server: SampleMCP

🧪 Testing Your MCP Servers

Method 1: Using MCP Inspector (Recommended)

# Make sure your virtual environment is activated first
.venv\Scripts\activate

# Test any of your servers
npx @modelcontextprotocol/inspector python mathserver.py
npx @modelcontextprotocol/inspector python weather.py
npx @modelcontextprotocol/inspector python workflow.py

This opens a web interface at http://localhost:5173 where you can:

• View available tools

• Test tool calls interactively

• Debug issues in real-time

Method 2: Direct Python Testing

# Run the server directly (will wait for stdin input)
python mathserver.py

# Or run tests
python test_mcp.py

📁 Project Structure

mcp_demo2/
├── .venv/                 # Virtual environment
├── .env                   # Environment variables (API keys)
├── .gitignore            # Git ignore file
├── requirements.txt      # Python dependencies
├── mathserver.py         # Math operations MCP server
├── weather.py           # Weather data MCP server  
├── workflow.py          # LangGraph workflow MCP server
├── test_mcp.py          # Test scripts
└── README.md            # This file

🔧 Available MCP Servers

1. Math Server (mathserver.py)

Simple arithmetic operations:

• add(a, b) - Add two numbers

• subtract(a, b) - Subtract two numbers

• multiply(a, b) - Multiply two numbers

Test example:

npx @modelcontextprotocol/inspector python mathserver.py
# Try: add(10, 5) → returns 15

2. Weather Server (weather.py)

Weather data operations (if implemented):

• Weather fetching tools

3. Workflow Server (workflow.py)

LangGraph-powered content generation:

• run_langgraph(input) - Generate jokes, stories, poems, or general responses

Test example:

npx @modelcontextprotocol/inspector python workflow.py
# Try: run_langgraph("tell me a joke about cats")

🔗 Integrating with Claude Desktop

1. Locate your Claude Desktop config:

   • Windows: %APPDATA%/Claude/claude_desktop_config.json

   • Mac: ~/Library/Application Support/Claude/claude_desktop_config.json

2. Add your MCP server:

   {
     "mcpServers": {
       "math": {
         "command": "python",
         "args": ["C:\\full\\path\\to\\mcp_demo2\\mathserver.py"],
         "env": {
           "PATH": "C:\\full\\path\\to\\mcp_demo2\\.venv\\Scripts;%PATH%"
         }
       },
       "workflow": {
         "command": "python", 
         "args": ["C:\\full\\path\\to\\mcp_demo2\\workflow.py"],
         "env": {
           "PATH": "C:\\full\\path\\to\\mcp_demo2\\.venv\\Scripts;%PATH%"
         }
       }
     }
   }

3. Restart Claude Desktop and test:

   • "What tools do you have available?"

   • "Add 15 and 25 for me"

   • "Tell me a joke about programming"

🐛 Troubleshooting

Common Issues:

"Module not found" error:

# Make sure virtual environment is activated
.venv\Scripts\activate
pip install fastmcp

Inspector shows empty tools:

• Check that your server file has @mcp.tool() decorators

• Verify the file runs without Python errors

• Make sure you're using transport="stdio" for Claude Desktop

Encoding errors (emojis in console):

• Remove emoji characters from print statements

• Or add UTF-8 encoding at the top of your Python files

Claude Desktop not finding tools:

• Use absolute paths in config file

• Make sure the virtual environment path is correct

• Restart Claude Desktop after config changes

📚 Learn More

• MCP Documentation

• FastMCP Documentation

• LangGraph Documentation

🤝 Contributing

1. Fork the repository

2. Create a feature branch

3. Make your changes

4. Test with MCP Inspector

5. Submit a pull request

📄 License

This project is open source. See LICENSE file for details.