NYC

hugging-face-datasets

Warn

Audited by Snyk on Feb 15, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.90). The skill's sql_manager.py explicitly uses DuckDB's hf:// protocol to load arbitrary Hugging Face Hub datasets (e.g., "cais/mmlu") and run SQL against them, meaning it fetches and interprets public, user-contributed dataset content from the open web (Hugging Face Hub) as part of its workflow.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 08:18 PM