A high-performance image cropping skill that enables AI agents to process local files or remote URLs into production-ready assets.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install crop-image
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 crop-image 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 Crop Image Skill?
The Crop Image skill is a specialized tool designed to empower AI agents with robust image manipulation capabilities. By integrating this capability within Openclaw Skills, developers can allow their agents to programmatically resize and trim images to specific dimensions. This service leverages a FastAPI backend to handle complex image processing tasks that would otherwise be difficult for an LLM to perform natively, providing a bridge between text-based reasoning and visual asset generation.
Crop Image Skill Use Cases
- Generating profile picture thumbnails from user-provided URLs in automated workflows.
- Processing local image assets during automated frontend development tasks within Openclaw Skills.
- Preparing media for social platforms requiring specific aspect ratios and dimensions.
- Normalizing dataset images for machine learning pipelines managed by AI agents.
- Automating the creation of UI assets by cropping large stock photos into specific component sizes.
How Crop Image Skill Works
- The AI agent identifies a need to crop an image based on user instructions or workflow requirements.
- The agent validates input parameters including the image source and desired dimensions which must be between 1 and 4096 pixels.
- A request is dispatched to the service endpoint, either via a JSON payload for remote URLs or multipart form-data for local file uploads.
- The image processing engine crops the source material and performs metadata analysis, such as detecting the presence of faces.
- The skill returns a structured JSON response containing the new production-ready URL and the original image dimensions.
- The agent uses the returned URL to complete the development or content creation task within the Openclaw Skills ecosystem.
Crop Image Skill Setup
To integrate this functionality into your environment, ensure your agent can communicate with the ImageClaw API. You can test the connectivity for URL-based cropping using the following command:
curl -sS -X POST "https://api.imageclaw.net/crop" \
-H "content-type: application/json" \
-d '{
"url": "https://picsum.photos/800/600",
"width": 256,
"height": 256
}'
For local file processing within Openclaw Skills, use the upload endpoint:
curl -sS -X POST "https://api.imageclaw.net/crop/upload" \
-F "file=@/absolute/path/to/photo.jpg" \
-F "width=256" \
-F "height=256"
Crop Image Skill Data Schema & Taxonomy
The Crop Image skill utilizes a structured data format to ensure compatibility across different Openclaw Skills implementations.
Input Parameters
| Parameter | Type | Description |
|---|---|---|
| url | string | Reachable image URL (for URL mode) |
| file | binary | Local image file (for upload mode) |
| width | integer | Target width between 1 and 4096 |
| height | integer | Target height between 1 and 4096 |
Response Object
cropped_url: The direct link to the resulting image asset.original_size: An array [width, height] of the source image.face_detected: A boolean flag indicating if human faces were identified during processing.
Crop Image Skill Advanced Features
- Smart face detection to help agents identify if subjects are centered.
- High-resolution support for professional-grade assets up to 4096px.
- Dual-mode operation supporting both cloud-based URLs and local file system paths.
- Detailed error reporting for invalid MIME types or unreachable network resources.
- Native compatibility with Claude Code and Claude Desktop for seamless Openclaw Skills integration.
SKILL.md
Loading
Related Openclaw Skills
Tier-1 VC Fundraising Agent Skill
An automated workflow for startups to evaluate venture-scale potential, craft investment narratives, and execute structured outreach to top-tier global investors.
Ma Xian'er Metaphysics Inference System
A professional-grade Chinese metaphysics engine providing deterministic calculations and AI-driven interpretations for Bazi, Ziwei Dou Shu, and divination.
AdvancedMLClassificationSkill
A comprehensive automation tool for generating industrial-grade machine learning classification pipelines, from data preprocessing to visual analytics.
WeChat Draft Publisher
A professional AI assistant that automates the entire WeChat Official Account publishing process from a single prompt.
Nightscout Local
A read-only integration for retrieving real-time CGM data and glucose trends from a Nightscout instance.
Molthouse
Molthouse is an API-first casino platform designed for AI agents to play provably fair games using USDC on the Base network.
Featured*
