Extract YouTube video metadata and transcripts into structured Markdown documents.
npx skills add https://github.com/vre/flow-state --skill youtube-to-markdown
Please help me install this Agent Skill. Skill: youtube-to-markdown Source: https://github.com/vre/flow-state/tree/main/youtube-to-markdown Command: npx skills add https://github.com/vre/flow-state --skill youtube-to-markdown
This Skill automates the extraction of YouTube video data, including metadata, description, and chapters, and prepares a structured Markdown footprint that you can drop into any knowledge vault.
Use the youtube-to-markdown skill to extract data for a YouTube URL like https://www.youtube.com/watch?v=example. The skill will generate youtube_{id}metadata.md, youtube{id}description.md, and youtube{id}_chapters.json in your output directory, ready for final assembly.
High-intent search queries and answers about installing and using this skill.
YouTube-to-markdown extracts video metadata, transcripts, descriptions, and chapters into a structured Markdown document. It outputs ready-to-use files containing video details, full transcription with timestamps, and chapter structure that you can import directly into Obsidian, Notion, or any note-taking system.
Yes. The Skill uses yt-dlp to retrieve video metadata, accessible subtitles, and chapter information, then assembles them into Markdown with proper heading hierarchy and timestamp preservation. Transcripts are deduplicated and segmented into paragraphs for readability.
The Skill extracts title, URL, uploader, view count, duration, description, and chapter timestamps. All metadata is formatted into structured Markdown files that preserve searchability and language information while maintaining a single consolidated document per video.
The Skill is designed for videos with accessible metadata, descriptions, chapters, and subtitles. Performance depends on subtitle availability; videos without captions will extract metadata and chapter data but cannot generate full transcripts.
Run YouTube-to-markdown on each video URL to generate individual Markdown files containing metadata, transcripts, and chapters. Collect the output files into a single folder structure, then import them into your note-taking system to create a compiled, searchable knowledge base organized by video.
Metadata extraction captures title, uploader, views, and duration for quick reference. Full transcription captures the complete spoken content with timestamps and paragraph breaks. YouTube-to-markdown performs both—metadata for discoverability and transcription for content capture and searchability.
glebisExtract YouTube transcripts and metadata into Markdown with YAML frontmatter.
JamesFincherRead, list, search, and create Markdown notes in an Obsidian vault.
vibecodeDownload videos, audio, and playlists from YouTube and other websites using yt-dlp.
ElliotJLTExtract YouTube video transcripts into timestamped Markdown files with metadata.
