What problem does it solve? Getting the content of a YouTube video normally requires watching it in full or relying on unofficial APIs with keys. This Skill fetches video subtitles directly through the InnerTube API without an API key, handles anti-bot fallbacks automatically, and turns the transcript into a structured summary matched to your intent. ## Core Features & Use Cases - Transcript Fetching: Downloads manual or auto-generated captions in multiple languages, with chapter segmentation, timestamps, translation, and local caching to avoid repeat network requests. - Intent-Based Summaries: Four output modes (quick overview, detailed chapter summary, tutorial analysis, targeted Q&A) selected from the user's request, using templates in references/summary-templates.md. - Speaker Identification: Optional mode that formats interview or dialogue videos with labeled speakers and chapter-segmented dialogue. - Use Case: Paste a YouTube link and ask "summarize this video" — the Skill fetches Chinese or English subtitles, caches them under .wopal-space/.tmp/, and returns a structured summary with key points and timestamps. ## Quick Start Summarize this YouTube video for me: https://www.youtube.com/watch?v=VIDEO_ID