The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes large external datasets which creates a surface for indirect prompt injection if the data contains malicious instructions.
Ingestion points: Instructions in SKILL.md describe loading data via vaex.open() and vaex.from_csv().
Boundary markers: No delimiters or 'ignore instructions' warnings are specified for the data being processed.
Capability inventory: The skill performs file reads/writes, statistical computations, and visualizations across the provided examples.
Sanitization: No validation or escaping of the data content is mentioned in the documentation.

vaex