transcribe-and-analyze

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly downloads and transcribes arbitrary public URLs (e.g., YouTube) via scripts/transcribe.py/yt-dlp (see SKILL.md and scripts/transcribe.py) and then analyze_transcript.py feeds the raw transcript text into LLM prompts (OpenAI/Ollama), so untrusted, user-generated third‑party content can directly influence model behavior and analysis.