dask
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructions direct the agent to ingest and process unstructured and semi-structured data from external sources such as CSV, Parquet, and JSON files. This creates a surface for indirect prompt injection where malicious instructions embedded in the processed data could attempt to influence the agent's logic.
- Ingestion points: The skill utilizes functions like
dd.read_csv(),db.read_text(), andda.from_zarr()to load external content into the agent's context (SKILL.md). - Boundary markers: No specific delimiters or instructions to ignore embedded commands within the data were identified in the provided examples.
- Capability inventory: The skill leverages Dask's distributed computing capabilities, including file system access and task execution across clusters via
dask.distributed. - Sanitization: The documentation does not explicitly detail sanitization or validation logic for the content of the files being processed.
Audit Metadata