youtube-transcript

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

• Third-party content exposure detected (high risk: 0.85). The required workflow fetches and ingests YouTube transcript caption text from an outsider source (public YouTube page/content via DeepAPI POST /v1/scrape/youtube/transcript and, on fallback, via yt-dlp producing *.json3 that is parsed into .txt), and that free-form text is then read into the agent/context for output/saving.