video-outpainting

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

• Third-party content exposure detected (high risk: 0.90). The skill accepts and downloads arbitrary source video URLs provided in the CLI input (see the "Invoke" examples with "video_url" and the "Security & Privacy" section noting "source video URLs are untrusted"), and those videos are ingested by the edit pipeline—so untrusted third-party content can be read/interpreted and materially influence model/tool behavior.