youtube-transcript-dossier

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

• Third-party content exposure detected (high risk: 0.85). The required workflow fetches a YouTube transcript (outsider-authored caption text) at runtime via scripts/fetch_transcript.py → youtube_transcript_api and then uses that transcript text to synthesize the dossier, so the LLM context can include arbitrary free text from the video.