hugging-face-datasets

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The documentation references running Python scripts via uv run scripts/dataset_manager.py and scripts/sql_manager.py. However, the skill explicitly states that these files are not bundled and are provided as historical integration designs. It includes clear warnings to not execute similarly named files from unrelated directories and to review any implementation before use.
  • [COMMAND_EXECUTION]: The skill provides examples of using duckdb for SQL queries and huggingface_hub for uploading data. These are standard operations for dataset management and follow best practices, including warnings against running arbitrary SQL or publishing sensitive data.
  • [DATA_EXPOSURE]: The skill provides guidance on using HF_TOKEN via a protected environment and emphasizes creating private repositories by default when uploading data.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 08:22 PM