data_analysis

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

• Third-party content exposure detected (medium risk: 0.30). Skill runtime ingests user-provided dataset content by loading arbitrary local files (e.g., via pl.read_csv/pl.read_json/pl.read_ndjson/pl.read_parquet or passing --file_path), then uses the parsed strings/numeric fields when generating reports and formatted output.