clipify-video-clip-generator

Generate social-ready vertical clips from long videos with face-tracking reframe and word-level captions.

4|Updated May 16, 2026
One-click install
npx skills add https://github.com/reason-machines/devtools-skills --skill clipify-video-clip-generator-reason-machines
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: clipify-video-clip-generator
Source: https://github.com/reason-machines/devtools-skills/tree/main/skills/clipify-video-clip-generator
Command: npx skills add https://github.com/reason-machines/devtools-skills --skill clipify-video-clip-generator-reason-machines

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai-whisper, numpy, ffmpeg, and includes scripts (resource) components.

What problem does it solve? Manually finding highlight moments in long videos, reframing horizontal footage to vertical, and adding styled captions is slow and repetitive. This Skill automates clip detection, 9:16 reframing, and subtitle burning using ffmpeg and Whisper. ## Core Features & Use Cases - Auto clip detection: Transcribes video with Whisper and proposes 3-5 clip candidates based on punchlines, reversals, audio peaks, and awkward pauses. - Face-tracking reframe: Converts 16:9 footage to 9:16 vertical using motion-energy-based speaker detection, without ML models. - Styled captions: Burns opus, karaoke, or minimal word-by-word subtitles via generated ASS files. - Use Case: Turn a 2-hour podcast interview into multiple TikTok-ready vertical clips with highlighted captions in about 20 seconds of render time per clip on Apple Silicon. ## Quick Start Invoke the /clipify command in Claude Code and provide the path to your source video to get proposed clips, then pick a clip, aspect ratio, and subtitle style.

Frequently Asked Questions about clipify-video-clip-generator

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

FAQPage Schema
How do I turn a long video into TikTok clips automatically?▼

Run the /clipify command with your source video path. It transcribes the video with Whisper, proposes 3-5 clip candidates with timestamps and reasons, then cuts, reframes to 9:16, and burns styled captions into the final MP4.

How to reframe 16:9 video to 9:16 vertical with two speakers?▼

Clipify generates motion energy files for each face region with ffmpeg crop filters, builds a speaker timeline via analyze.py, then creates an ffmpeg crop pan expression with build_pan.py that follows the active speaker.

Does Clipify work on Linux or Windows?▼

Yes, but you must remove the macOS-specific -hwaccel videotoolbox flag from the ffmpeg commands in SKILL.md. Linux can use CUDA acceleration and Windows can use DXVA2, or run CPU-only encoding.

Why are my video captions out of sync with Whisper?▼

Whisper timestamps drift on long videos. Transcribe only the relevant short segment instead of the full file, or enable word-level timestamps with --word_timestamps True for more accurate caption alignment.

What subtitle styles does Clipify support?▼

Three built-in styles: opus (bold white text with yellow active-word highlight), karaoke (word-by-word color change at the bottom), and minimal (clean white text). You can also edit the generated ASS file for custom fonts, colors, and animations.