MCP Jupyter Complete

MCP Jupyter Complete is an MCP server that provides tools for manipulating Jupyter notebook files through position-based cell operations and VS Code synchronization. It connects Model Context Protocol clients directly to local .ipynb files, allowing data scientists, machine learning practitioners, and Python developers to programmatically inspect, modify, and restructure notebooks. Users can view cell listings, retrieve specific code or markdown content, edit existing cells, insert new cells at designated indices, and delete cells with automatic reindexing. The server also supports bulk modifications, converting cells between code, markdown, and raw types, and moving cells across positions. Additionally, it provides specific VS Code integration tools, such as triggering editor reloads, to ensure external edits made by automated agents reflect immediately in active IDE sessions without file desynchronization or corrupting notebook metadata.

Category: Data & Analytics

Tags: data-science, jupyter, notebooks, python, vscode

Visit MCP Jupyter Complete

How to install and configure MCP Jupyter Complete

1. Clone the repository and install dependencies: bash git clone https://github.com/tofunori/mcp-jupyter-complete.git cd mcp-jupyter-complete npm install 2. Optionally run npm link to make the executable available globally. 3. Add the server configuration to your ~/.claude.json file: json { "mcpServers": { "jupyter-complete": { "command": "node", "args": ["/path/to/mcp-jupyter-complete/src/index.js"] } } } If installed globally via npm, you can specify "command": "mcp-jupyter-complete" with an empty argument list. 4. Restart your MCP client to start using the tools.

What you can do with MCP Jupyter Complete

• Listing all notebook cells by index and type to understand notebook structure before applying programmatic changes. - Editing existing cell code or markdown content directly via specific cell indices without modifying raw JSON by hand. - Inserting newly generated Python data analysis scripts or markdown documentation blocks at specific notebook positions. - Converting cell types between markdown, raw, and code when adjusting notebook formatting or narrative structure. - Executing bulk cell updates in a single call to refactor multi-step notebook workflows consistently. - Triggering VS Code file reloads automatically after agent edits to keep active development environments in sync.

Key facts

• Open Source
• https://github.com/tofunori/mcp-jupyter-complete
• Data & Analytics, Developer Tools & Code Intelligence
• data-science, jupyter, notebooks, python, vscode

Part of MCP Servers

What is MCP Jupyter Complete?

MCP Jupyter Complete is a Model Context Protocol server designed for programmatic manipulation of Jupyter notebook files. It allows AI assistants to read, edit, delete, insert, and convert notebook cells using explicit zero-based index positions, while providing utility triggers to synchronize changes directly with active VS Code notebook editors.

How do I install MCP Jupyter Complete?

Clone the repository using git clone, navigate to the directory, and run npm install. You can optionally link the package globally using npm link. Then, configure your client by adding the node execution command pointing to src/index.js in your claude.json configuration file.

What cell types are supported?

MCP Jupyter Complete supports code cells for executable Python scripts, markdown cells for formatted explanatory text, and raw cells for unformatted text. You can create, edit, or convert cells between any of these three supported formats.

How does VS Code integration work?

The server includes a trigger_vscode_reload tool that forces VS Code to reload modified notebook files. When combined with Claude Code and recommended VS Code extensions like Python and Jupyter, external changes made through MCP calls prompt the editor to update without manual file reloading.

Is MCP Jupyter Complete open source?

Yes, MCP Jupyter Complete is open-source software released under the MIT License. The complete source code, documentation, and licensing terms are publicly available on GitHub.