The Agent Skills Directory

[Command Execution] (LOW): The skill utilizes jupyter nbconvert --execute to run code contained within .ipynb files. This is the primary function of the skill, but executing untrusted notebooks can lead to arbitrary code execution on the host system.
[Indirect Prompt Injection] (LOW): The skill processes Jupyter notebook files which are external data sources that may contain malicious instructions.
Ingestion points: The skill reads notebook.ipynb using json.load().
Boundary markers: No delimiters or warnings are used to prevent the agent from obeying instructions embedded in notebook cells.
Capability inventory: The skill performs file system writes, searches via grep, and code execution via nbconvert.
Sanitization: There is no validation or sanitization of notebook cell content prior to execution.

jupyter