hot-notebook-patching 0.1.0

JupyterLab extension for hot patching externally modified notebooks

hot-notebook-patching

A JupyterLab extension for hot patching notebooks that are edited externally — for example by an AI coding agent like Claude Code.

Instead of triggering a full notebook reload (which restarts the kernel and clears widget state), only the cells that actually changed are patched into the live notebook model.

What it does

Claude Code edits notebook.ipynb on disk
          ↓
Extension detects the file change
          ↓
Only modified cells are patched
          ↓
Kernel keeps running · Widget state intact · Scroll position preserved

Modified lines are highlighted with an orange left-border marker directly in the cell editor, similar to git-diff gutter decorations in VS Code.

Requirements

• JupyterLab ≥ 4.0
• Notebooks must use nbformat 4.5+ (cell IDs required for accurate diffing)

Installation

pip install hot-notebook-patching

Usage

Toolbar button

Each notebook gets a ⚡ bolt button in the toolbar. Click it to toggle auto-patch mode — the extension will watch the file for changes and patch automatically.

Commands

Action	How
Patch once from disk	Toolbar ⚡ button (when auto-patch is off) · Shift+P inside a notebook · Command Palette → Patch Notebook From Disk
Toggle auto-patch	Toolbar ⚡ button · Command Palette → Toggle Auto-Patch on File Change

What is preserved

• Running kernel
• Widget state (ipywidgets, anywidget, …)
• Cell outputs and execution counts
• Scroll position

What is updated

• Cell source code (with line-level diff indicators)
• Cell metadata
• Added / deleted / reordered cells

Conflict detection

If both you and an external tool edited the same cell since the last patch, the extension skips that cell and logs a warning — it will never silently overwrite your changes.

Auto-execute on patch

Add "hotpatch": true to a cell's metadata and the cell will be automatically executed after its source is patched:

{
  "hotpatch": true
}

Development

git clone https://github.com/kolibril13/hot-notebook-patching
cd hot-notebook-patching

# Create venv and install JupyterLab
uv venv .venv
uv pip install --python .venv/bin/python jupyterlab

# Install JS dependencies and build
.venv/bin/jlpm install
.venv/bin/jlpm build:lib:prod
.venv/bin/jlpm build:labextension:dev

# Install Python package in editable mode
uv pip install --python .venv/bin/python hatchling hatch-jupyter-builder editables
uv pip install --python .venv/bin/python -e . --no-build-isolation

# Link the extension for live development
.venv/bin/jupyter labextension develop . --overwrite

# Start JupyterLab
.venv/bin/jupyter lab

For iterative development, rebuild after source changes:

.venv/bin/jlpm build:lib:prod && .venv/bin/jlpm build:labextension:dev

Then hard-refresh JupyterLab (Cmd+Shift+R / Ctrl+Shift+R).

License

MIT