hugging-face-datasets

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 datasets (e.g., "cais/mmlu" and full hf:// paths) — public, user-contributed dataset content is fetched and then read/interpreted by the agent via SQL queries/samples, exposing it to untrusted third‑party input that could carry indirect prompt injection.