Productivity
youtube-transcribe-skill avatar

youtube-transcribe-skill

Extract YouTube video subtitles or transcripts directly into local text files using yt-dlp or browser automation.

Introduction

The youtube-transcribe-skill is a specialized automation tool designed for researchers, content creators, and developers who need to quickly convert YouTube video content into searchable or editable text. By integrating directly into your development workflow, it eliminates the manual labor of transcribing audio or copying and pasting from web interfaces. The skill prioritizes efficiency by first attempting a high-speed command-line extraction, then seamlessly falling back to browser-based UI automation if restricted or missing required CLI tools.

  • Automatically detects and supports various YouTube URL formats, including standard watch, shareable youtu.be, and short-form video URLs.

  • Utilizes yt-dlp to efficiently fetch subtitle tracks, auto-generated captions, and metadata, ensuring minimal resource consumption.

  • Implements a robust fallback mechanism using Chrome DevTools MCP, allowing it to navigate YouTube's dynamic DOM and interact with transcript panels when CLI methods encounter network or permission constraints.

  • Cleans raw output by stripping VTT/SRT styling tags, headers, and duplicate timestamps, providing a clean, human-readable plain text result in a consistent Timestamp-Text format.

  • Supports multi-language subtitle extraction, enabling the retrieval of Chinese (Simplified/Traditional) and English captions where available.

  • To ensure optimal performance, verify that the yt-dlp command-line tool is installed on your system path. If yt-dlp is unavailable, ensure Chrome DevTools MCP is configured to enable the browser automation fallback mode.

  • Input requirements consist of a valid YouTube URL provided via command arguments or inferred from the active conversation context.

  • Output is automatically saved to the local working directory using the video title as the filename in .txt format, making it ideal for downstream tasks like summarization, indexing, or documentation.

  • Always ensure you have appropriate browser cookies accessible (defaulting to Chrome) if the video has age restrictions or regional sign-in requirements, as the skill leverages these for authenticated access.

Repository Stats

Stars
1,483
Forks
218
Open Issues
0
Language
Python
Default Branch
main
Sync Status
Idle
Last Synced
May 3, 2026, 05:11 PM
View on GitHub