An automated workflow for transforming long-form videos into high-engagement, captioned short-form clips for social media.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install clips-machine
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install clips-machine using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Clips Machine?
Clips Machine is a professional-grade utility designed to bridge the gap between long-form content and viral social media presence. By utilizing Openclaw Skills, this tool automates the tedious process of manual video editing, allowing creators to repurpose podcasts, streams, and interviews into high-impact vertical videos for platforms like TikTok, Instagram Reels, and YouTube Shorts.
This skill is built on a foundation of 100% free, local tools, ensuring that your data remains private and your costs remain zero. It handles everything from the initial download and transcription to AI-driven virality analysis and final rendering with trendy, animated captions. It is an essential asset for anyone looking to scale their content production without a dedicated editing team.
Clips Machine Use Cases
- Automatically extracting the most engaging segments from 2-hour podcasts for social media teasers.
- Converting long Twitch VODs into funny gaming highlights with meme-style captions.
- Generating professional LinkedIn video snippets from corporate interviews or webinars.
- Creating a high-frequency posting schedule for TikTok by batch-processing existing video archives.
- Identifying key emotional peaks and quotable moments within raw footage for marketing reels.
How Clips Machine Works
- Provide a video source, such as a local MP4 file or a URL from YouTube, TikTok, or Twitter.
- The system uses yt-dlp to ingest the media and Whisper to generate a high-accuracy, timestamped transcript locally.
- An AI analysis engine scans the transcript to identify hooks, emotional peaks, and controversial takes, assigning each a virality score.
- The best segments are selected based on the user-defined clip count or the highest scores.
- FFmpeg reformats the chosen segments into a 9:16 vertical aspect ratio and applies dynamic, animated captions.
- The final assets, including the video clips and a summary report, are saved to a structured directory for immediate distribution.
Clips Machine Setup
To deploy this skill, you must have the required dependencies installed on your machine. These Openclaw Skills rely on local binaries for maximum performance and privacy.
# For macOS users
brew install ffmpeg yt-dlp whisper-cpp
# For Linux users
sudo apt install ffmpeg
pip install yt-dlp
# Build whisper.cpp from source to ensure local compatibility
After installation, ensure these tools are in your system PATH so the agent can invoke them correctly.
Clips Machine Data Schema & Taxonomy
Clips Machine organizes all generated data within a specific project folder located at ~/Videos/OpenClaw/clips-[video-name]/. The following structure ensures all metadata and media are easily accessible:
| File | Purpose |
|---|---|
transcript.json |
Full JSON transcription containing all words and their specific timestamps. |
viral_moments.json |
A technical breakdown of detected clips, including start/end times and virality scores. |
clip_n.mp4 |
The final rendered vertical video files with hardcoded captions. |
summary.md |
A human-readable overview of the processing results and clip descriptions. |
Clips Machine Advanced Features
- Support for multiple viral captioning styles including the high-conversion Hormozi style, karaoke, and minimal layouts.
- Local AI virality detection that evaluates hooks, emotional peaks, and actionable advice without cloud API costs.
- Comprehensive source support, allowing you to pull content from YouTube, TikTok, X (Twitter), and Twitch VODs.
- Batch processing capabilities to generate up to 10 or more clips from a single long-form video in one command.
- Seamless integration with other Openclaw Skills to facilitate automated social media management pipelines.
SKILL.md
Loading
Related Openclaw Skills
ohmyopenclaw
A comprehensive collection of AI-native configuration guides designed to automate the setup and optimization of Openclaw Skills for diverse agent architectures.
Homey Smart Home Control
A comprehensive CLI-based skill for controlling Athom Homey smart home devices through local LAN or cloud APIs.
casedev
A comprehensive legal AI platform providing encrypted document storage, OCR, transcription, and specialized legal search capabilities.
Memory Maintenance
An intelligent memory management tool that reviews daily notes and automates file cleanup for OpenClaw agents.
YouTube Factory
A comprehensive automation skill that transforms a single text prompt into a complete YouTube video with voiceover, footage, and captions.
AgentMail
AgentMail is an API-first email infrastructure that allows AI agents to create programmatic inboxes, send high-volume messages, and process incoming mail via real-time webhooks.
METADATA
Featured*
