gentic-data

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests public, user-provided data via create_table_from_csv (HTTPS/S3/Google Sheets/Drive links) and then reads/samples that data with sample_table/get_table_schema and uses it to drive updates/syncs, so untrusted third‑party content can materially influence agent actions.