What is Mcp Server Poc

mcp-server-poc is a proof-of-concept implementation of a Model Context Protocol (MCP) server designed to enhance AI assistant capabilities by providing custom tools and resources, including simple math operations, dynamic greetings, and web crawling functionalities.

Use cases

Use cases for mcp-server-poc include providing AI assistants with access to documentation resources, performing web searches, and executing simple mathematical operations to assist users in real-time.

How to use

To use mcp-server-poc, clone the repository, set up a Python 3.11 virtual environment, install the required packages, configure environment variables, and run the application using ‘python main.py’. Integration with Cursor IDE is also supported through a specific configuration file.

Key features

Key features of mcp-server-poc include a documentation search tool for libraries like LangChain and OpenAI, web crawling capabilities using the crawl4ai library, and integration with the Google Search API.

Where to use

mcp-server-poc can be used in various fields where AI assistants are deployed, particularly in applications requiring enhanced information retrieval and processing capabilities.

MCP Note Taker (POC)

A proof-of-concept implementation of a Model Context Protocol (MCP) server for AI assistant note-taking, featuring custom tools and resources for managing notes.

Overview

This project demonstrates how to create and use a Model Context Protocol (MCP) server that provides note-taking capabilities to AI assistants (such as Claude, Cursor, and others supporting MCP). The server includes:

• A tool to add notes
• A resource to fetch the latest note
• A prompt to summarize all notes

All notes are stored in a local notes.txt file in the project directory.

Requirements

• Python 3.11
• Required packages listed in requirements.txt

Installation and Setup

1. Clone this repository:

git clone https://github.com/yourusername/mcp-server-poc.git
cd mcp-server-poc

2. Create and activate a virtual environment:

# Init uv package manager 
uv init

# Create a Python 3.11 virtual environment
uv venv

# Activate on Windows
 .venv\Scripts\activate

# Activate on macOS/Linux
source venv/bin/activate

3. Install the required packages:

uv pip install -r requirements.txt

4. (Optional) Install MCP CLI tools if needed for development or alternative integrations:

uv add "mcp[cli]"

Running the Application

To run the MCP server:

uv run mcp

The server will start and wait for connections using the stdio transport method.

Integrating with Cursor

To use this MCP server with Cursor IDE:

1. Create or edit the file ~/.cursor/mcp.json (on Windows: C:\Users\<username>\.cursor\mcp.json) with the following content:

{
  "mcpServers": {
    "mcp-server": {
      "command": "python",
      "args": [
        "ABSOLUTE/PATH/TO/main.py"
      ]
    }
  }
}

2. Replace the path with the absolute path to your main.py file.

   • On Windows, use double backslashes: C:\\Users\\username\\path\\to\\main.py
   • On macOS/Linux, use regular slashes: /Users/username/path/to/main.py

3. Restart Cursor completely (including ending any background processes) to load the MCP server.

Features

Tools

• add_note(note: str): Adds a note to the notes.txt file and returns a confirmation message.

Resources

• notes://latest: Returns the latest note from the notes.txt file, or a message if there are no notes yet.

Prompts

• note_summary_prompt(): Generates a prompt asking the AI to summarize all current notes in notes.txt.

Technical Details

• All notes are stored in a plain text file named notes.txt in the project root. This file is created automatically if it does not exist.
• The server uses the Model Context Protocol SDK and the mcp[cli] dependency.

Windows Binary Mode Fix

If you use stdio transport on Windows, you may need to set binary mode for stdin/stdout. See the MCP documentation for details.

Troubleshooting

• Verify that all required packages are installed (pip list to check)
• Check that the absolute path in the configuration file is correct
• Make sure the MCP server is running with the proper version of Python (3.11)
• If you encounter issues, try running the MCP server directly to see any error output

License

MIT License

Acknowledgements

• This project uses the Model Context Protocol SDK