Jupyter Notebook

Jupyter Notebook

Use this skill to create clean, reproducible notebooks for two primary modes:

• experiments and exploratory analysis
• tutorials and teaching-oriented walkthroughs

When to use

• Create a new notebook from scratch.
• Convert rough notes or scripts into a structured notebook.
• Refactor an existing notebook to be more reproducible and skimmable.
• Build experiments or tutorials that other people will re-run or read.

Decision tree

• If the request is exploratory or hypothesis-driven, choose experiment.
• If the request is instructional or step-by-step, choose tutorial.
• If editing an existing notebook, preserve intent and improve structure.

Quick start

If this skill is installed under DeepAgents, replace ~/.cline/skills in the examples below with ~/.deepagents/agent/skills.

Use the bundled helper:

python "$HOME/.cline/skills/jupyter-notebook/scripts/new_notebook.py" \
  --kind experiment \
  --title "Compare prompt variants" \
  --out output/jupyter-notebook/compare-prompt-variants.ipynb

python "$HOME/.cline/skills/jupyter-notebook/scripts/new_notebook.py" \
  --kind tutorial \
  --title "Intro to embeddings" \
  --out output/jupyter-notebook/intro-to-embeddings.ipynb

Workflow

1. Lock the intent and audience.
2. Scaffold from the helper instead of hand-authoring notebook JSON.
3. Fill the notebook with small, runnable steps.
4. Use references/experiment-patterns.md or references/tutorial-patterns.md as needed.
5. Validate top-to-bottom execution when the environment allows.
6. If execution is not possible, say so explicitly and describe local validation.

Bundled resources

• Templates: assets/experiment-template.ipynb, assets/tutorial-template.ipynb
• Structure guidance: references/notebook-structure.md
• Final checklist: references/quality-checklist.md

Dependencies

The scaffold script uses only the Python standard library.

Optional packages for local execution:

python -m pip install jupyterlab ipykernel