Video Resize for Openclaw

A local and AI-powered video reformatting tool designed to convert any video into platform-specific aspect ratios.

symbolk
v1.0.0
Feb 28, 2026
1
1.1k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install video-resize

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install video-resize 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 Video Resize?

The Video Resize skill is a versatile utility within the Openclaw Skills ecosystem that automates the process of changing video aspect ratios. It allows developers and creators to transform horizontal footage into vertical formats for social media or vice versa without manual editing. By leveraging local ffmpeg processing, it ensures fast center-cropping with zero latency or API costs.

For more complex scenarios, this Openclaw Skills extension integrates with an AI Edit tool to provide subject-aware cropping. This advanced mode identifies the most important elements in a frame—such as a speaker in an interview—and intelligently adjusts the crop to keep them centered. This makes it an essential tool for high-volume content repurposing across platforms like TikTok, Instagram, and YouTube.

Video Resize Use Cases

  • Converting landscape YouTube videos into vertical 9:16 clips for TikTok and Shorts.
  • Resizing standard footage into 1:1 square format for Instagram posts.
  • Reformatting cinematic 21:9 ultrawide content into 16:9 for standard displays.
  • Using AI subject-tracking to reframe talking-head interviews for mobile-first vertical viewing.
  • Batch processing video assets for cross-platform digital marketing campaigns.

How Video Resize Works

  1. The user specifies the source video file and the desired target aspect ratio (e.g., 9:16, 1:1).
  2. The skill determines if a local center-crop is sufficient or if AI-powered subject tracking is required.
  3. For local tasks, the skill runs an ffmpeg command to crop and scale the video to standard resolutions like 1080x1920.
  4. For AI tasks, the video is uploaded to a secure processing API where it undergoes frame analysis based on selected style tips.
  5. The skill monitors the processing status and provides the final file path or a temporary download URL for the processed asset.

Video Resize Setup

To use the local resizing features of this Openclaw Skills entry, ensure ffmpeg is installed on your system:

# For macOS
brew install ffmpeg

# For Linux
sudo apt install ffmpeg

To enable subject-aware AI cropping, configure your API credentials:

openclaw config set env.SPARKI_API_KEY "your_api_key_here"

You can then run a basic resize command immediately:

bash scripts/resize.sh video.mp4 "9:16"

Video Resize Data Schema & Taxonomy

The skill manages video metadata and file outputs according to the following schema:

Property Type Description
Input File Path Path to the source .mp4 or .mov file.
Ratio String Supported: 9:16, 1:1, 16:9, 4:3, 21:9.
Resolution String Automatically calculated (e.g., 1080x1920 for 9:16).
Style Tips Integer AI identifiers for style (1=Energetic, 4=Talking-head).
Result URL String Temporary download link for AI-processed videos.

Video Resize Advanced Features

  • Subject-aware AI recomposition that intelligently follows the speaker in the frame.
  • High-performance local processing using H.264 and AAC encoding for maximum compatibility.
  • Support for multiple cinematic styles including Fast-paced, Cinematic, and Highlight Reel.
  • Automated workflow polling with built-in timeout handling and error recovery.
  • Custom user prompts to guide the AI editing process for specific creative directions within the Openclaw Skills framework.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*