yt-structure

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

• Third-party content exposure detected (high risk: 0.85). The required workflow ingests outsider-authored free text by downloading YouTube auto-captions from a user-supplied URL (public web content) via yt-dlp, then converting the resulting .vtt caption text into a .txt that is subsequently used for LLM ingestion.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches arbitrary external content at runtime (yt-dlp on user-supplied YouTube URLs and the curl "" → pandoc pattern) which is injected into the LLM context and can directly control prompts — e.g. https://youtube.com/watch?v=abc123