Jupythunder logo

Jupythunder

by kbsooo

Jupythunder is a middleware that connects the LLM chat interface (Claude) with Jupyter Notebook. This project allows users to interact with Jupyter Notebook in a way that's closer to natural language.

View on GitHub

Last updated: N/A

What is Jupythunder?

Jupythunder is a middleware that bridges the gap between LLM chat interfaces like Claude and Jupyter Notebook, enabling natural language interaction with Jupyter.

How to use Jupythunder?

To use Jupythunder, you need to clone the repository, install the dependencies, run the Jupyter Notebook server, run the Jupythunder server, configure Claude for Desktop to connect to the Jupythunder server, and then interact with Jupyter using natural language commands through Claude.

Key features of Jupythunder

  • Natural Language Code Conversion

  • Jupyter Remote Connection

  • Code Execution

  • Code Questions

  • State Management

  • Notebook File Management

  • Natural Language Data Analysis

  • Cell Query and Execution

Use cases of Jupythunder

  • Natural language data analysis and visualization

  • Automated code generation and execution within Jupyter

  • Interactive exploration of Jupyter notebooks using natural language commands

  • Remote access and control of Jupyter notebooks through chat interfaces

FAQ from Jupythunder

How do I resolve the 'could not connect to MCP server jupythunder: spawn pythonENOENT' error?

This error occurs when the Python executable file cannot be found. Use the absolute path to the Python executable in the claude_desktop_config.json file or use the python3 command.

What do I do if Jupyter connection fails?

Check if the Jupyter server is running, verify that the provided URL and token are correct, and check firewall settings.

What causes notebook creation or opening failures?

Check if you have write permissions to the Jupyter server, make sure the notebook name does not contain special characters, and check if a notebook with the same name already exists.

What should I do if I encounter cell execution issues?

Check for syntax errors in the code, verify that necessary variables or functions are defined in previous cells, and restart the server if execution results are not displayed in the notebook.

What are the future improvements planned for Jupythunder?

Future improvements include LLM integration, data visualization extensions, notebook templates, file upload support, collaboration features, interactive widgets, and GPU acceleration.