MCP Server

1. Create virtual environment and activate it

   uv venv
   source .venv/bin/activate
   

2. Install dependencies

   uv add mcp httpx arxiv
   

3. To launch it by running

   uv --directory /ABSOLUTE/PATH/TO/PARENT/FOLDER/mcp run server.py
   

4. Create .env file

   touch .env
   

5. Add your key to the .env file:

   ANTHROPIC_API_KEY=<your_key_here>
   

6. To test your MCP Server you can use mcp inspector

   npx @modelcontextprotocol/inspector uv run server.py
   

MCP Client

1. Install required packages

   uv add mcp anthropic python-dotenv nest_asyncio
   

2. To run your client with any MCP server:

   uv run client.py path/to/server.py # python server
   uv run client.py path/to/build/index.js # node server
   

   Relative path

   uv run client.py ./server/server.py
   

   Absolute path

   uv run client.py /Users/username/projects/mcp/server.py
   

   Windows path (either format works)

   uv run client.py C:/projects/mcp/server.py
   uv run client.py C:\\projects\\mcp\\server.py
   

Deployment

1. If the server using pip instead of uv

   uv pip compile pyproject.toml > requirements.txt
   

2. Make sure the server use the right python version

   echo "python-3.11.11" > runtime.txt
   

3. Input the required environment variables

4. Run the mcp server

   python server.py
   

references :

• https://www.youtube.com/watch?v=wa_A0qY0anA
• https://learn.deeplearning.ai/courses/mcp-build-rich-context-ai-apps-with-anthropic/lesson/fkbhh/introduction