Skip to content
Skill

youtube-transcripter

by TylonHH

AI Summary

Extracts and summarizes YouTube video transcripts into structured notes with key learnings and actionable suggestions. Useful for researchers, students, and content consumers who want quick, clean summaries without manual transcription.

Install

Copy this and paste it into Claude Code, Cursor, or any AI assistant:

I want to install the "youtube-transcripter" skill in my project.

Please run this command in my terminal:
# Install skill into the correct directory
mkdir -p .claude/skills/codex_skill_youtube-transcripter && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/codex_skill_youtube-transcripter/SKILL.md "https://raw.githubusercontent.com/TylonHH/codex_skill_youtube-transcripter/main/SKILL.md"

Then restart Claude Code (or reload the window in Cursor) so the skill is picked up.

Description

Extract a YouTube video transcript from a URL and summarize it into important content, learnings, and suggestions. Use when the user provides a YouTube link and wants a clean transcript and structured notes without timestamps, with fallback extraction methods if the primary tool is unavailable.

Overview

Extract a transcript from a YouTube URL and produce a concise, structured summary focused on key content, learnings, and actionable suggestions.

1) Confirm input

• Require a YouTube URL or video ID. Multiple inputs are allowed. • Assume no timestamps unless the user explicitly asks. • Default language is auto-detected; ask only if a specific language is required.

2) First-time setup (one-time)

Make sure at least one transcript extractor is installed. Prefer yt-dlp: `bash choco install yt-dlp -y ` If Chocolatey is not available or not elevated, use pip: `bash python -m pip install yt-dlp ` Optional (API-based): `bash python -m pip install youtube-transcript-api `

3) Extract transcript (best-effort with fallback)

Preferred: run the bundled script: `bash python scripts/get_transcript.py "https://www.youtube.com/watch?v=VIDEO_ID" ` Multiple inputs (URLs or IDs) into a folder: `bash python scripts/get_transcript.py VIDEO_ID_1 "https://youtu.be/VIDEO_ID_2" --out-dir transcripts ` Notes: • --out is only for single input. • --out-dir writes transcript_<id>.txt for each input. Default is transcripts/. • --wrap 125 reflows lines to 125 chars; use --wrap 0 to keep raw line breaks. To force a language: `bash python scripts/get_transcript.py "https://www.youtube.com/watch?v=VIDEO_ID" --lang en --out transcript_VIDEO_ID.txt ` Fallback order if the script fails: 1) Install yt-dlp (prefer Chocolatey if available) Check: `bash choco -v ` If Chocolatey is available, install (requires admin shell): `bash choco install yt-dlp -y ` If Chocolatey is not available or not elevated, use pip: `bash python -m pip install yt-dlp ` 2) yt-dlp with auto captions (binary or Python module) `bash yt-dlp --skip-download --write-auto-sub --sub-lang en --sub-format vtt "https://www.youtube.com/watch?v=VIDEO_ID" ` Then strip timestamps and cue markers from the .vtt file. 3) If tools are unavailable, ask the user for the transcript text or a downloaded caption file.

Discussion

0/2000
Loading comments...

Health Signals

MaintenanceCommitted 2mo ago
Active
AdoptionUnder 100 stars
2 ★ · Niche
DocsREADME + description
Well-documented

GitHub Signals

Stars2
Issues0
Updated2mo ago
View on GitHub
MIT License

My Fox Den

Community Rating

Sign in to rate this booster

Works With

Claude Code