Turn YouTube videos into content

Start with the words spoken in a video and build useful written outputs without inventing the missing visuals. This skill includes a Python standard-library helper that invokes a separately installed yt-dlp to retrieve public captions only, plus editorial workflows for summaries, chapters, threads, blogs, newsletters, and carousels. Timestamped segments preserve an evidence trail. Prompt templates explain chunking, attribution, uncertain transcription, and format-specific structure. Missing dependencies, unavailable caption tracks, and access failures produce actionable errors. No API key is required, and offline fixtures exercise parsing and caption-only command construction. Captions remain dependent on public availability; the helper does not download audio or video or automatically write the content drafts.