The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data sources which could contain malicious instructions designed to influence the agent. 1. Ingestion points: Data is loaded via dd.read_csv, db.read_text, and da.from_zarr as described in SKILL.md and references/dataframes.md. 2. Boundary markers: No explicit prompt boundaries or instructions to ignore data-embedded commands are provided. 3. Capability inventory: The skill has access to Bash(python:*), Bash(uv:*), and file Write/Edit tools, which are high-capability. 4. Sanitization: No content-level sanitization for the processed datasets is mentioned.
[COMMAND_EXECUTION]: The skill utilizes Bash tools to execute Python-based Dask workflows and manage clusters, which is standard for its intended purpose.

alterlab-dask