Jupyter MCP Server

Jupyter MCP Server is a Model Context Protocol (MCP) server that enables interaction with Jupyter notebooks running in a local JupyterLab environment. It facilitates real-time collaboration (RTC) through Jupyter's features.

To start, install jupyterlab, jupyter-collaboration, and ipykernel. Launch JupyterLab with a specified port, token, and IP address, ensuring accessibility for the MCP server, especially when running in a Docker container.

For integration with Claude Desktop, configure the claude_desktop_config.json file with the server URL, token, and notebook path, adjusting the Docker run command based on the operating system (MacOS, Windows, or Linux).

The server provides tools like add_execute_code_cell (executes code), add_markdown_cell (adds markdown), and download_earth_data_granules (downloads Earth data).

Build the server using Docker. Alternatively, install it for Claude Desktop via Smithery using the provided command.