analyzing-video

Extract frames, detect scene changes, and transcribe audio into a Markdown report.

24|5|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/bsisduck/video-analyzer-skill --skill analyzing-video
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analyzing-video
Source: https://github.com/bsisduck/video-analyzer-skill/tree/main
Command: npx skills add https://github.com/bsisduck/video-analyzer-skill --skill analyzing-video

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ffmpeg, ffprobe, python3, bc, whisper, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates video analysis by extracting frames, detecting scene changes, transcribing audio, and compiling a structured Markdown report.

Core Features & Use Cases

  • Frame extraction and montage grids for quick visual summaries
  • Scene-change detection with high-resolution key frames for important moments
  • Parallel subagents for grid visuals, key frames, and audio transcription
  • Transcription options: auto, user-provided, or visual-only
  • Output: final Markdown analysis document with timeline and scene breakdown
  • Use Case: analyze social media clips, tutorials, and marketing videos to generate summaries and highlight moments.

Quick Start

Analyze a sample video: /path/to/video.mp4

Frequently Asked Questions about analyzing-video

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

FAQPage Schema
How do I automatically extract frames and transcribe audio from a video?

To automatically extract frames and transcribe audio from a video, this Skill orchestrates ffmpeg and whisper to process video files, detect scene changes, and compile a structured Markdown report. It handles both visual frame extraction and audio transcription in parallel.

Do I need ffmpeg and whisper installed to analyze video content?

Yes, you need ffmpeg and whisper installed to analyze video content, along with python3, ffprobe, and bc. These dependencies are required for executing frame extraction, scene-change detection, and audio transcription commands.

Can I analyze long marketing videos or only short social media clips?

You can analyze long marketing videos and short social media clips. The Skill applies automatic tier selection to handle varying video lengths, allowing custom frame rates and time ranges to process both short and long-form content effectively.

What is the best way to generate a visual summary with scene-change detection?

The best way to generate a visual summary with scene-change detection is using an automated video analysis pipeline. This Skill detects scene changes, creates montage grids for quick visual summaries, and extracts high-resolution key frames for important moments.

Does video analysis work without an audio transcript?

Video analysis works without an audio transcript by offering visual-only extraction options. You can choose auto transcription, provide a user-provided transcript, or skip audio entirely to focus solely on frame extraction and scene detection.

Why does my video frame extraction produce a Markdown report?

Video frame extraction produces a Markdown report because the Skill synthesizes extracted frames, scene-change data, and audio transcriptions into a structured final document. This output provides a timeline and scene breakdown for easy review.