henry0hai/simple-mcp-server

Simple MCP Server & Client Example

This project demonstrates how to build and interact with a simple Model Context Protocol (MCP) server using Python. The server exposes tools and resources, and the client shows how to call them asynchronously.

Features

• Add two numbers (add tool)
• Search Google using SerpAPI (search_google tool)
• Get server config (get_config resource)
• Get a personalized greeting (personalized_greeting resource)

Setup

1. Clone the repository

git clone git@github.com:henry0hai/simple-mcp-server.git
cd simple-mcp-server

2. Create and activate a Python virtual environment (optional but recommended)

python3 -m venv mcp-env
source mcp-env/bin/activate

3. Install dependencies

pip install -r requirements.txt

4. Set up environment variables

Create a .env file in the project root with your SerpAPI key:

SERPAPI_KEY=your_serpapi_key_here

Running the MCP Server

python server.py

The server will start on http://127.0.0.1:8000/mcp (or the port set in your .env as MCP_SERVER_PORT).

Running the MCP Client

python client.py

You should see demo calls to the add tool, get_config resource, and personalized_greeting resource, with results printed to the console.

Adding More Tools

• To add more tools, define a new function in server.py and decorate it with @mcp.tool().
• For new resources, use @mcp.resource("resource://your_resource").

Notes

• The search_google tool requires a free SerpAPI key.
• For production use, secure your API keys and consider error handling and rate limits.

Notes: I using qdrant as a vector database for storing and retrieving embeddings.

docker run -p 6333:6333 -p 6334:6334 \
    -v "$(pwd)/qdrant_storage:/qdrant/storage:z" \
    qdrant/qdrant

License

MIT