video-review

Extract transcripts and keyframes from online videos for summarization.

2|2|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/codysumpter-cloud/buddy-brain --skill video-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: video-review
Source: https://github.com/codysumpter-cloud/buddy-brain/tree/main/skills/video-review
Command: npx skills add https://github.com/codysumpter-cloud/buddy-brain --skill video-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires yt-dlp, ffmpeg, whisper, and includes scripts (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the review and summarization of online videos by extracting transcripts and keyframes, providing a quick and efficient way to understand the content of videos.

Core Features & Use Cases

  • Video Summarization: Automatically summarize the content of a video using extracted transcripts and keyframes.
  • Transcription Extraction: Extract and use transcripts from videos for analysis and summary.
  • Keyframe Extraction: Sample keyframes from videos to visually represent important moments.
  • Use Case: If you receive a video link and need a quick summary, this Skill can analyze the video and provide a summary with timestamps, saving you time from watching the entire video.

Quick Start

To review a video and summarize its content, run the command: python3 skills/video-review/scripts/review_video.py --url "<video-url>" --out-dir "./video-review".

Frequently Asked Questions about video-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I summarize a YouTube video without watching the whole thing?

To summarize a YouTube video, this Skill downloads the content, extracts transcripts and keyframes, and generates a text summary with timestamps. This provides a quick review of online videos without watching the entire duration.

Do I need ffmpeg and whisper installed to extract video transcripts?

Yes, ffmpeg and whisper are required dependencies for video transcript extraction. You also need yt-dlp to download the online video content before the transcription and keyframe extraction processes can run.

Can I get a video summary with timestamps for research workflows?

Yes, you can get a video summary with timestamps for research workflows. The Skill processes online video links to extract keyframes and transcripts, outputting a summarized document ideal for administrative and research analysis.

What is the best way to extract keyframes from an online video?

The best way to extract keyframes from an online video is using an automated script that samples visual frames alongside audio transcripts. This Skill uses ffmpeg to sample keyframes, providing a visual representation of important video moments.

How do I run the video review and summarization script from the command line?

To run the video review and summarization script, execute `python3 skills/video-review/scripts/review_video.py --url "<video-url>" --out-dir "./video-review"`. This command processes the specified video and outputs the summary files.

Why does video transcription fail without yt-dlp?

Video transcription fails without yt-dlp because the Skill relies on it to download the online video content first. Without the downloaded media file, ffmpeg and whisper cannot process the video to extract transcripts and keyframes.