This MCP server provides a lightweight environment to manage sticky notes via the Model Context Protocol. It exposes three core capabilities as MCP tools: add_note(message) to append a new note to sticky_notes.txt, read_notes() to retrieve all notes, and get_latest_note() to fetch only the most recent entry. Prompts are available to generate a summary of all notes (note_summary_prompt). You can connect this server to GitHub Copilot in agent mode or to Claude AI, enabling conversational workflows where the AI models create, view, and summarize notes directly through natural language prompts. When running, Copilot or Claude AI can invoke add_note, read_notes, and get_latest_note to interact with the backing text file, enabling use cases like quick note taking during conversations and automatic summarization of prior entries.

Prerequisites:

• Python 3.10 or higher
• uv (Python virtual environment/manager) installed or available in PATH
• Git installed (optional, for cloning repositories)
• Access to a shell/terminal (PowerShell, CMD, Bash, etc.)

Step-by-step installation:

1. Clone the repository (or use your existing project structure):

   git clone https://github.com/ingfcetina/mcp-servers.git cd mcp-servers

2. Set up a Python environment (recommended):

   Create a virtual environment (example using venv)

   python -m venv .venv

   Activate the environment:

   Windows (PowerShell)

   ..venv\Scripts\Activate.ps1

   macOS/Linux

   source .venv/bin/activate

3. Install UV and MCP CLI dependencies:

   pip install uv[mcp]

   Optional additional dependencies mentioned in the project

   pip install pydantic python-dotenv pyyaml

4. Ensure your server script exists and is accessible:

   • Place your main server script at project/main.py (or adjust the path in the mcp_config)

5. Run the MCP server (as shown in the documentation):

   uv run --with mcp mcp run project/main.py

6. Verify the Inspector MCP UI opens at http://localhost:6277 (default port) for development and testing.

• The mcp_config example uses uv to run the MCP server with the MCP CLI integration. Adjust the path to your main.py as needed for your project structure.
• If you encounter port conflicts (6277), change the Inspector port or disable it during testing.
• Ensure your environment variables (if any) are set correctly, especially when integrating with Claude AI or Copilot agents.
• For Windows paths in configuration (e.g., Claude or VS Code settings), prefer forward slashes or proper escaping in JSON configuration.
• The provided server focuses on a simple text-based notes store; extend or replace main.py to add more complex data models or storage backends if required.