two-sigma-ml-at-scale

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md Alternative Data Pipeline (ingest_satellite_imagery/fetch_images and the "web scraping" mention) explicitly fetches and processes external/public data which is untrusted and is then read/interpreted to compute features and drive backtests/models, so third-party content can materially influence agent decisions.