youtube-analysis

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches transcripts from public YouTube pages using scripts/fetch_transcript.py (youtube-transcript-api and yt-dlp) and then instructs Claude to "read the full transcript" and perform the analysis (SKILL.md Step 4 and scripts/analyze_video.py appends transcript_text), so untrusted, user-generated content from YouTube is ingested and directly drives the agent's decisions/outputs.