youtube-content

Extract transcripts and metadata from YouTube videos for structured analysis.

1|Updated Nov 11, 2012
One-click install
npx skills add https://github.com/fairchild/dotfiles --skill youtube-content-fairchild
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: youtube-content
Source: https://github.com/fairchild/dotfiles/tree/main/agents/shared/first-party-skills/youtube-content
Command: npx skills add https://github.com/fairchild/dotfiles --skill youtube-content-fairchild

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires youtube-transcript-api, yt-dlp, pyyaml, and includes scripts (resource) and references (resource) components.

What problem does it solve? Analyzing YouTube videos requires manually watching content or copy-pasting captions. This Skill fetches transcripts and metadata programmatically, then applies structured analysis templates so you can summarize, extract insights, or pull quotes from any public video. ## Core Features & Use Cases - Transcript and Metadata Fetching: Retrieves captions via youtube-transcript-api and metadata (title, channel, duration, views) via yt-dlp, output as JSON. - Five Analysis Modes: Wisdom extraction, summary, Q&A generation, quote extraction with timestamps, and custom analysis templates. - Knowledge Base Persistence: Automatically saves analyses as Markdown with YAML frontmatter and raw transcripts as JSON, with search and tag filtering. - Use Case: Paste a conference talk URL and ask for key insights—the Skill fetches the transcript, extracts actionable takeaways and notable quotes, and saves the analysis to your searchable knowledge base. ## Quick Start Ask the AI to summarize this YouTube video and extract the key insights: https://youtube.com/watch?v=VIDEO_ID

Frequently Asked Questions about youtube-content

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract a transcript from a YouTube video in Python?▼

Use youtube-transcript-api to fetch captions by video ID, returning text and optional timestamped segments. The fetch_youtube.py script wraps this with yt-dlp metadata extraction and outputs combined JSON.

How to get YouTube video metadata without downloading the video?▼

Use yt-dlp with skip_download enabled to extract title, channel, duration, view count, upload date, and tags. The script runs yt_dlp.YoutubeDL with extract_info and download disabled.

What YouTube URL formats are supported for transcript extraction?▼

Supported formats include youtube.com/watch?v=ID, youtu.be/ID, youtube.com/embed/ID, youtube.com/v/ID, URLs with timestamps, and bare 11-character video IDs. The script parses these with regex patterns.

Why does transcript extraction fail for some YouTube videos?▼

Extraction fails when captions are disabled, the video is private or deleted, or no transcript exists. The script returns partial results with metadata and an errors array describing the specific failure.

Can I search previously analyzed YouTube videos?▼

Yes, analyses are saved to a knowledge base directory with YAML frontmatter. Use search_knowledge.py to list recent entries, search by keyword across titles and content, or filter by tags.