What it does
Use when the user asks to create, scaffold, or edit Jupyter notebooks (.ipynb) for experiments, explorations, or tutorials; prefer the bundled templates and run the helper script new_notebook.py to generate a clean starting notebook.
This entry summarizes the public skill for discovery. Prefer the upstream SKILL.md for the full instruction text (we do not paste restricted or lengthy proprietary prompt bodies).
Install & usage
Codex
Inside Codex, use the skill installer (repo is deprecated as a catalog but curated skills remain usable):
$skill-installer jupyter-notebook
Or copy from GitHub:
mkdir -p .codex/skills
git clone --depth 1 https://github.com/openai/skills.git /tmp/openai-skills
cp -R /tmp/openai-skills/skills/.curated/jupyter-notebook .codex/skills/jupyter-notebook
Restart Codex after install so it discovers the skill.
Best for
Scaffolding clean experiment/tutorial notebooks.
Hands-on test
- test_level: fully-tested
- Ran: yes
- When: 2026-11-01T01:20:00+08:00 (Asia/Shanghai)
- Result: Created/edited an .ipynb and executed cells via nbclient/jupyter locally.
- Gotchas: Kernel deps must exist in the venv; empty kernels fail silently without checks.
Money-saving tip
Prefer ChatGPT Go ($8/mo US) or Plus ($20/mo) before Pro tiers; Business annual Standard seats are $20/user/mo vs $25 monthly. Nonprofit Standard seats: $8 annual / $10 monthly (help.openai.com).
Source & license
- source_repo: openai/skills
- Source: https://github.com/openai/skills/tree/main/skills/.curated/jupyter-notebook
- License: Apache-2.0 — https://github.com/openai/skills/blob/main/skills/.curated/jupyter-notebook/LICENSE.txt
- Restrictions: Individual skill LICENSE.txt applies (usually Apache-2.0; vercel-deploy is MIT). openai/skills catalog is deprecated — prefer openai/plugins for newer examples, but these curated skills remain useful.