skills.rest
JulesDups
JulesDupsCOMMUNITY
Explore all skills by this author →
GitHub

analyze-video

Download videos, transcribe audio locally, extract frames, and generate per-video syntheses with a cross-video recap.

Updated May 22, 2026
Search Tags:#automation#multi-agent#video#transcription#frame-extraction#markdown-synthesis
One-click install
npx skills add https://github.com/JulesDups/claude-config-public --skill analyze-video-julesdups
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: analyze-video
Source: https://github.com/JulesDups/claude-config-public/tree/main/skills/analyze-video
Command: npx skills add https://github.com/JulesDups/claude-config-public --skill analyze-video-julesdups

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires yt-dlp, ffmpeg, ffprobe, faster_whisper, ctranslate2, and includes scripts (resource) components.

What problem does it solve?

This skill automates the end-to-end workflow of downloading videos from multiple sources, transcribing locally when possible, extracting frames for visual context, and dispatching parallel sub-agents to produce per-video syntheses plus a final cross-video recap.

Core Features & Use Cases

  • Download videos from YouTube, Vimeo and direct mp4 URLs.
  • Transcribe audio locally using faster-whisper when GPU is available.
  • Extract frames at configurable intervals to support visual synthesis.
  • Dispatch parallel sub-agents to write one Markdown synthesis per video and generate an aggregate recap.
  • Silent orchestration with strict cleanup and a final reporter recap.

Quick Start

Run the analyze-video skill with one or more video URLs to start download, transcription, and synthesis.

Frequently Asked Questions about analyze-video

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

FAQPage Schema
How do I automate video transcription and frame extraction for multiple URLs?▼

You can automate video transcription and frame extraction by passing multiple URLs to this skill, which downloads the videos, extracts frames at configurable intervals, and transcribes audio locally using faster-whisper.

Does local video transcription with faster-whisper require a GPU?▼

Local video transcription with faster-whisper utilizes GPU acceleration when available, but processes offline using yt-dlp, ffmpeg, and ctranslate2 to download and transcode media before generating Markdown syntheses.

What is the best way to generate a cross-video recap from multiple video sources?▼

Generating a cross-video recap is handled by dispatching parallel sub-agents to write individual Markdown syntheses per video, followed by an aggregate recap that summarizes the entire batch.

Can I download and transcribe videos from Vimeo and direct mp4 URLs offline?▼

Yes, you can download and transcribe videos offline from Vimeo, YouTube, and direct mp4 URLs, as the skill uses yt-dlp for fetching and applies strictly silent orchestration with comprehensive cleanup.

Are there limitations to batch processing video sources with modular agent-based synthesis?▼

Batch processing video sources with modular agent-based synthesis requires local dependencies like ffmpeg and ctranslaate2, limiting execution to environments that support offline execution and optional GPU hardware.

Related Skills

Explore More Software Engineering Skills→
PlaneInABottlePlaneInABottle

youtube-downloader

Download YouTube videos or extract audio to MP3 with yt-dlp.

Community
Intermediate
jg-chalk-iojg-chalk-io

moai-cc-claude-md

Automates Markdown documentation generation and publication workflows for Claude Code.

Community
Advanced
iliaaliliaal

orchestrating-swarms

Orchestrate multi-agent teams with leader-teammate models and JSON messaging.

Community
Advanced
video-production-buddyvideo-production-buddy
602

video-understand

Analyze local videos with ffmpeg frame extraction and Whisper transcription.

Official
Intermediate