working-in-notebooks
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill documents the use of shell commands within notebooks (
!ls,!pip) and through Python'ssubprocessmodule. These are standard features for environment setup and repository management in data science workflows. - [EXTERNAL_DOWNLOADS]: The documentation provides instructions for installing well-known data science packages and developer tools (e.g., Jupyter, marimo, uv, poetry) from official registries and GitHub repositories.
- [REMOTE_CODE_EXECUTION]: The guide covers Jupyter magic commands (
%run,%%writefile) and marimo's reactive execution model. These are core functionalities of the respective notebook environments intended for interactive development. - [SAFE]: The skill includes explicit security guidance, specifically advising against hardcoding API keys and providing instructions for using environment variables and secret managers (AWS, GCP). It also advocates for the use of
.gitignoreandnbstripoutto prevent accidental data leaks.
Audit Metadata