Skip to content
TakoAPI
Back to skills

youtube-video-analyzer

by sdrabent

Coding Agents & IDEs
640Downloads reported by clawskills.sh
1Stars
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Multimodal YouTube video analysis through both audio (transcript) and visual (frame extraction + image analysis)

Description

What This Skill Does

Analyzes YouTube videos through both transcript extraction and frame-by-frame visual analysis. Synchronizes spoken content with extracted video frames by timestamp, enabling step-by-step guides where on-screen UI elements, code, and diagrams are captured alongside what is said.

Unlike transcript-only YouTube tools, it captures visual content, so information shown on screen but not spoken, such as specific button names, menu paths, or code, is not lost.

When to Use It

  • Building step-by-step guides from software tutorial videos
  • Extracting code shown on screen but not read aloud
  • Documenting exact UI click paths from product demo videos
  • Summarizing long explainer videos with timestamped visual anchors
  • Capturing tool usage and hand positions from physical howto videos

Example Workflow

INPUT

User asks: Create a step-by-step guide from this YouTube tutorial: https://youtube.com/watch?v=abc123

AGENT
  1. 1

    Fetches video metadata (title, duration, ID) using yt-dlp

  2. 2

    Extracts timestamped transcript via yt-dlp JSON dump and curl

  3. 3

    Downloads video at 720p and extracts keyframes at adaptive intervals using ffmpeg

  4. 4

    Synchronizes each frame with the corresponding transcript segment by timestamp

  5. 5

    Reads each frame image alongside transcript text to generate a structured step-by-step guide

OUTPUT

A markdown step-by-step guide with each action timestamped and annotated with visual descriptions of what appears on screen at that moment

Requirements

  • ffmpeg
  • yt-dlp

Installation

clawhub install sdrabent/youtube-video-analyzer

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links