fal-recipes

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

• Third-party content exposure detected (high risk: 0.80). The recipes explicitly accept and ingest user-supplied public URLs for images/videos (e.g., "Source image: URL or local path" in references/image-restoration.md and character-lipsync.md) and the genmedia workflow uses those external image_url/video_url inputs as reference material driving prompt construction and endpoint selection, so arbitrary third‑party content can directly influence tool actions and prompts.