Skill Detail

Detect Scene Boundaries And Export Cut Lists Before Video Review With Pyscenedetect

Detect scene changes in a video, emit cut boundaries, and hand back machine-usable scene lists before manual review, splitting, or clip assembly.

Media & TranscriptionMulti-Framework
Media & Transcription Multi-Framework Security Reviewed
โญ 4.7k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill detect-scene-boundaries-and-export-cut-lists-before-video-review-with-pyscenedetect Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Python 3, pip, ffmpeg or mkvmerge for splitting support
Install & setup
pip install scenedetect[opencv] –upgrade
Author
Brandon Castellano and contributors
Publisher
Open Source
Last updated
Apr 17, 2026
Quick brief

PySceneDetect is a narrow, operator-facing skill for finding scene boundaries in video files and turning them into cut lists, saved images, or split-ready segments. An agent should invoke it when a downstream workflow needs reliable scene timestamps before someone starts reviewing footage by hand, extracting clips, or chunking media for transcription.

How it works

What this skill actually does

Use this instead of a full video editor when the real job is scene detection and cut-list generation, not timeline editing. The scope boundary is explicit: detect scene transitions, export scene metadata, and optionally hand off to ffmpeg or another step for splitting. It is not a general media-production or video-generation listing.