mcp-server-jupyter
The MCP server enables programmatic management and interaction with Jupyter notebooks. It offers tools for reading, editing, and executing notebook cells, making it useful for integrating Jupyter functionalities into other platforms such as Claude Desktop.
What is the purpose of the MCP Server for Jupyter?
The MCP Server for Jupyter allows programmatic management and interaction with Jupyter notebooks, facilitating automation and integration into workflows.
How can I add a new cell to a notebook?
Use the 'add_cell' tool, specifying the notebook path, source content, and optionally the cell type and position.
Can I read the output of a specific cell?
Yes, use the 'read_output_of_cell' tool by providing the notebook path and the cell ID.
Is it possible to execute a cell programmatically?
Yes, the 'execute_cell' tool allows you to execute a specific cell and retrieve its output.
Do I need to keep JupyterLab running while using Claude Desktop?
Yes, you need to keep JupyterLab or Jupyter Notebook running to interact with notebooks through Claude Desktop.