jupyter-live-kernel

To maintain state across Python code executions in a Jupyter environment, use a stateful Python REPL with a live Jupyter kernel. This setup preserves variables and program state incrementally, enabling iterative code development without losing prior computation results.

A stateful Python REPL is used for iterative data science exploration, allowing you to execute code incrementally and analyze datasets while preserving environment state. It supports complex analysis tasks by retaining variable values across multiple code executions.

Yes, you need JupyterLab installed to use a live Python kernel for iterative coding. The stateful REPL requires a JupyterLab setup and Python environments with specified libraries to execute interactive code and preserve state.

To execute Python code in an existing notebook using a live Jupyter kernel, use the command `uv run "$SCRIPT" execute --path <notebook.ipynb> --code '<python code>' --compact`. This sends code directly to the notebook for interactive execution.

Yes, you can discover available Jupyter servers and notebooks programmatically by running `uv run "$SCRIPT" servers --compact`. This command scans the environment to identify active servers and accessible notebooks for stateful REPL execution.

A live Jupyter kernel works better than a standard Python REPL for complex analysis because it integrates directly with JupyterLab and preserves state across executions. This stateful behavior allows data scientists to iteratively test code without reloading environments.