product-video

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

• Third-party content exposure detected (high risk: 0.70). Yes — the workflow explicitly requires collecting image assets from public, user-contributed sites (Phase 2: "收集图片素材 (见 asset-collection skill)" and the Technical Stack listing "Unsplash/Pexels | Pixabay"), so the agent will fetch and ingest untrusted third‑party content as part of its required processing, which can influence asset selection and downstream composition decisions.