Notebook Metadata
Documentsnotebook-metadataNotebook Metadata is an Agent Skill for documents published in the albumentations-team/albumentations_examples repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Ensures Jupyter notebooks have required metadata and nbformat. Use when adding, creating, or modifying .ipynb files, or when the user asks about notebook structure or validation.
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/albumentations-team/albumentations_examples/blob/HEAD/.cursor/skills/notebook-metadata/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/notebook-metadata, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
Notebook Metadata and nbformat
When to Apply
- Adding or creating new Jupyter notebooks
- Modifying notebook structure
- User asks about notebook validation or metadata
Required Structure
Every .ipynb file must have these top-level keys:
| Key | Type | Notes |
|---|---|---|
metadata | object | Can be empty {} but must exist |
nbformat | int | Must be 4 or 5 |
nbformat_minor | int | Minor version |
cells | list | Array of cell objects |
Minimal Valid Notebook
{
"cells": [],
"metadata": {},
"nbformat": 4,
"nbformat_minor": 4
}
Validation
Run the project script:
python scripts/check_notebook_metadata.py [path]
Omit path to check current directory recursively. Pre-commit runs this automatically on staged .ipynb files.
Fixing Invalid Notebooks
If a notebook is missing fields, add them at the root level:
import json
with open("notebook.ipynb") as f:
nb = json.load(f)
nb.setdefault("metadata", {})
nb.setdefault("nbformat", 4)
nb.setdefault("nbformat_minor", 4)
nb.setdefault("cells", [])
with open("notebook.ipynb", "w") as f:
json.dump(nb, f, indent=1)