content-cascade

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text because it fetches the YouTube transcript for a user-supplied or auto-detected video URL and then uses that transcript as the source material for generation (via Apify or youtube-transcript/yt-dlp), meaning arbitrary speech/poison in the video can reach the LLM.