Yt Video Analyzer banner
Auxlife32 Auxlife32

Yt Video Analyzer

AI community

Description

A Claude Code skill that turns a YouTube video into a digest with timestamps and interface screenshots. Subtitles and frames run locally; letting a model watch the video is optional.

Installation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open the source below and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

README

yt-video-analyzer

**A [Claude Code](https://claude.com/claude-code) skill that turns a YouTube video into a digest with timestamps and interface screenshots.**

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE) [![Python 3.8+](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/) [![Claude Code Skill](https://img.shields.io/badge/Claude%20Code-skill-d97757.svg)](https://claude.com/claude-code) [![X](https://img.shields.io/badge/X-@auxlifes-000000.svg?logo=x)](https://x.com/auxlifes) [![Telegram](https://img.shields.io/badge/Telegram-vicshot-2AABEE.svg)](https://t.me/vicshot)

**English** · [Русский](README.ru.md)

Drop in a link to a tutorial and get back not a summary, but a **map of techniques**: which tool was shown at 0:38, a screenshot of that exact moment, and an explanation of how to do it yourself.


How it works

Pipeline: link → subtitles → frames → model watches → digest

Three levels, cheapest first. The first two need no API key and no paid service — only `yt-dlp` and `ffmpeg` on your machine.

Level What you get Cost
Subtitles 80% of the meaning. No subtitles? whisper kicks in free
Frames a screenshot of the exact moment a button was shown free
Model watches what text can't show: cut rhythm, transition style ≈ $0.08 per 15 min

What it's actually for

**A tutorial on Seedance 2.5 → a working video-generation skill.**

A 16-minute walkthrough went through this tool. What came back wasn't a summary — it was the prompt block structure, the exact timing formats, the reference-tagging syntax and the generation settings, all read off the screen rather than out of the narration. That digest went to Claude with one request: *build me a skill that works this way.