Repolode
Submit
op7418

op7418/Youtube-clipper-skill

A Claude Code skill that downloads YouTube videos, splits them into AI-generated chapters, clips segments, and burns bilingual subtitles into the output.

View on GitHub
Stars
2,245
Forks
319
Open issues
12
Last commit
9 months ago

What is it?

YouTube Clipper Skill is a Claude Code skill for turning long YouTube videos into shorter clips. It downloads a video with yt-dlp, has Claude read the English subtitles to propose topic-based chapters of roughly 2 to 5 minutes, and then lets you choose which chapters to cut.

For each selected chapter, FFmpeg extracts the segment, translates the subtitles into a bilingual English and Chinese SRT file in batches, and can hardcode those subtitles into the video. It also generates social media summaries for Xiaohongshu, Douyin and WeChat. Outputs are organized into timestamped folders under a configurable output directory.

Setup requires Python 3.8+, yt-dlp and FFmpeg built with libass for subtitle burning. The project can be installed with npx skills or a bundled install script, and configured through a .env file.

Key features

  • 01Generates topic-based chapters of 2 to 5 minutes using Claude's reading of the subtitles.
  • 02Clips video segments with FFmpeg.
  • 03Batch translates subtitles into bilingual Chinese and English SRT files.
  • 04Burns customizable subtitles into the video.
  • 05Creates social media summaries for Xiaohongshu, Douyin and WeChat.
  • 06Downloads videos up to a configurable resolution limit.

Use cases

  • Pulling highlight clips from long interviews or podcasts for social media.
  • Turning lecture videos into shorter bilingual clips with subtitles.
  • Producing Chinese-subtitled versions of English YouTube content.

Quick start

  1. 1

    Install the skill with npx

    npx skills add https://github.com/op7418/Youtube-clipper-skill
  2. 2

    Install yt-dlp (macOS)

    brew install yt-dlp
  3. 3

    Install FFmpeg with libass (macOS)

    brew install ffmpeg-full
  4. 4

    In Claude Code, ask it to clip a YouTube video by giving the URL

Things to know

  • Requires Claude Code and system tools yt-dlp and FFmpeg with libass installed.
  • macOS users must replace the standard ffmpeg package with ffmpeg-full to burn subtitles.

Topics

#claude-code#video-editing#ffmpeg#subtitles#yt-dlp#translation#ai-agent

FAQ

What is Youtube-clipper-skill?
A Claude Code skill that downloads YouTube videos, splits them into AI-generated chapters, clips segments, and burns bilingual subtitles into the output. YouTube Clipper Skill is a Claude Code skill for turning long YouTube videos into shorter clips. It downloads a video with yt-dlp, has Claude read the English subtitles to propose topic-based chapters of roughly 2 to 5 minutes, and then lets you choose which chapters to cut.
Is Youtube-clipper-skill free to use? What license does it have?
Youtube-clipper-skill is open source under the MIT license. Check the license terms before using it in commercial products.
Is Youtube-clipper-skill still maintained?
No commits for 6+ months. The last commit to Youtube-clipper-skill was on Jan 22, 2026.
What can I use Youtube-clipper-skill for?
Pulling highlight clips from long interviews or podcasts for social media. Turning lecture videos into shorter bilingual clips with subtitles. Producing Chinese-subtitled versions of English YouTube content.
How do I install Youtube-clipper-skill?
Install the skill with npx: npx skills add https://github.com/op7418/Youtube-clipper-skill Install yt-dlp (macOS): brew install yt-dlp Install FFmpeg with libass (macOS): brew install ffmpeg-full In Claude Code, ask it to clip a YouTube video by giving the URL
Where can I get Youtube-clipper-skill?
The source code and installation instructions are on GitHub: https://github.com/op7418/Youtube-clipper-skill

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects