A focused Model Context Protocol server for safely reading, editing, and executing one live Jupyter notebook.
Robust new jupyter server is an MCP project maintained in the kshitijdesai99/jupyter_mcp_server_kshitij repository. The maintainer describes its purpose as “A focused Model Context Protocol server for safely reading, editing, and executing one live Jupyter notebook”. Read the project documentation for the complete scope.
The project provides a local client configuration using uv. Follow the repository’s build and installation steps before copying it, and replace example paths with the paths on your own computer.
The example declares NOTEBOOK_PATH, SERVER_URL, TOKEN. Secret-looking values have been replaced with placeholders. Required and optional labels are shown only when the documentation states them.
Project documentation · Repository · Discovery listing. This profile records the public documentation checked on 2026-10-08.
{
"mcpServers": {
"jupyter": {
"command": "uv",
"args": [
"--directory",
"/absolute/path/to/jupyter_mcp_server_kshitij",
"run",
"jupyter-mcp"
],
"env": {
"NOTEBOOK_PATH": "notebook.ipynb",
"SERVER_URL": "http://localhost:8888",
"TOKEN": "YOUR_TOKEN"
}
}
}
}From the project documentation. Replace placeholders and example paths before connecting.
| Variable | Description | Required |
|---|---|---|
| NOTEBOOK_PATH | Declared in the maintainer’s client configuration. | Check documentation |
| SERVER_URL | Declared in the maintainer’s client configuration. | Check documentation |
| TOKEN | Declared in the maintainer’s client configuration. | Check documentation |
Give your assistant a real code sandbox — isolated cloud VMs for actually running the code it writes.
Runs Python code locally within an interactive REPL environment.
Executes Python code within ephemeral Daytona sandboxes, facilitating tasks such as code execution, file manipulation, and web server previewing.
Enables Claude to manage software development projects with project context awareness and code execution via Docker.
Give your agent a persistent workspace: files, a code sandbox and a public URL to publish to.
Build Apps and run code in 30 languages — sandboxed, with persistent sessions for agent loops.