yt_dlp

Extract media metadata, transcripts, and audio from YouTube and other sites.

1|Updated Mar 11, 2026
One-click install
npx skills add https://github.com/antonyfmunoz/OS --skill yt-dlp-antonyfmunoz
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: yt_dlp
Source: https://github.com/antonyfmunoz/OS/tree/main/skills/tools/yt_dlp
Command: npx skills add https://github.com/antonyfmunoz/OS --skill yt-dlp-antonyfmunoz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

yt-dlp removes the need to manually manage YouTube/video downloads and transcription inputs by letting you programmatically fetch metadata, audio, and subtitles from a URL.

Core Features & Use Cases

  • YouTube research without API keys: search via ytsearchN:query and pull structured metadata as JSON using metadata-only runs.
  • Transcript and subtitle acquisition: fetch auto-generated captions (VTT) without downloading the full video for downstream cleaning and analysis.
  • Audio extraction for speech-to-text pipelines: download and convert audio to MP3 to feed Whisper/Groq-style transcription workflows.
  • Controlled format selection and rate safety: choose bestaudio/bestvideo+bestaudio and reduce throttling risk with retry/sleep/limit-rate options.

Quick Start

Ask your AI to search YouTube for "AI automation" and return JSON metadata plus English auto-transcripts for the top 5 results using yt-dlp.

Frequently Asked Questions about yt_dlp

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

FAQPage Schema
How do I download YouTube transcripts and subtitles without downloading the full video?▼

You can fetch YouTube transcripts and subtitles without downloading the video using write-auto-subs with skip-download, which acquires auto-generated VTT captions for downstream cleaning and analysis.

Can I extract YouTube metadata in bulk for search and research workflows?▼

Yes, you can extract YouTube metadata in bulk using ytsearchN:query to search and pull structured JSON metadata for top results without needing any API keys.

What's the best way to extract audio from YouTube videos for speech-to-text pipelines?▼

The best way to extract audio for speech-to-text pipelines is using extract-audio to download and convert audio to MP3, requiring ffmpeg, to feed Whisper or Groq-style transcription workflows.

Does this YouTube downloader require API keys to fetch video metadata?▼

No, this YouTube downloader does not require API keys to fetch video metadata; it uses a CLI/library downloader to retrieve structured JSON metadata directly from URLs.

Can I download audio and transcripts from playlists and channels, not just single videos?▼

Yes, you can download audio, transcripts, and metadata from playlists and channels, as well as single videos, applying to caption pipelines and audio extraction across multiple video sources.

How do I prevent throttling when downloading media from URLs at scale?▼

You can prevent throttling when downloading media at scale by applying retry, sleep, and limit-rate options alongside controlled format selection to reduce throttling risk during extraction.