A powerful 4-in-1 image processing toolbox for AI agents, featuring OCR, watermark removal, smart image compression, and professional ID photo generation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install toolbox-10-skill
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 toolbox-10-skill 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 Image Utility Toolbox?
The Image Utility Toolbox is a comprehensive local image processing suite designed to empower AI agents with robust computer vision and editing operations. Built using Python, this skill bundles four high-frequency image workflows: precise watermark elimination, multilingual OCR text extraction, intelligent size compression, and standard ID photo formatting with automated background replacement.
By integrating this tool, developers can bypass expensive cloud-based APIs and execute complex visual tasks directly in their secure local environments. This skill is fully optimized for Openclaw Skills configurations, enabling large language models to naturally orchestrate and run python-based scripts on image assets with perfect accuracy and file safety.
Image Utility Toolbox Use Cases
- Automated Text Extraction (OCR): Quickly scan and parse mixed Chinese and English text from diagrams, documents, or screenshots, returning exact layout coordinates.
- Watermark Erasure: Programmatically detect and paint over unwanted text or image watermarks automatically or via coordinates.
- Smart Size Minimizer: Compress JPEG/PNG assets using targeted quality algorithms or specified maximum file sizes without degrading necessary visual fidelity.
- ID Photo Color & Size Swapping: Instant background removal and resizing for official documents, supporting red, blue, or white background replacements.
How Image Utility Toolbox Works
- Trigger Recognition: The user issues natural language commands like "read text from image" or "convert this selfie to a blue background ID photo".
- Parameter Preparation: The AI agent extracts relevant file paths and determines optional parameter values, such as target dimensions or compression rates.
- Local Script Dispatch: The agent executes the specific python sub-script (e.g.,
remove_watermark.pyorid_photo.py) using absolute directory paths. - Structured JSON Output: The script analyzes the image, performs processing via Open Source packages, and outputs a structured JSON response details which the agent prints cleanly for the user.
Image Utility Toolbox Setup
Ensure you have python3 installed on your machine. Install the required Python packages before running the scripts:
python3 -m pip install easyocr opencv-python-headless Pillow rembg numpy onnxruntime
Note: On the first run of the OCR tool, easyocr will automatically attempt to fetch language models (approx. 100-300MB). Keep an active internet connection open or pre-define your model directories using the EASYOCR_MODULE_PATH environment variable.
Image Utility Toolbox Data Schema & Taxonomy
Script Inventory
| Utility | Executable Python Script | Key Operational Capability |
|---|---|---|
| Watermark Eraser | scripts/remove_watermark.py |
Auto-detects watermarks or heals precise bounding box areas (--x1 --y1 --x2 --y2). |
| OCR Reader | scripts/ocr_recognize.py |
Detects Chinese/English text and prints corresponding positions. |
| Smart Compressor | scripts/compress_image.py |
Compresses formats down to specific --quality (30-95) or maximum sizes in KB. |
| ID Passport Maker | scripts/id_photo.py |
Extracts human subjects to overlay on --bg colors (red, blue, white). |
Parameter Configuration
--auto: Triggers automatic inpainting mode for watermarks.--bg <color>: Sets ID photo background. Supported options:red,blue(default), orwhite.--quality <30-95>: Manually specifies target quality ratio (defaults to85).--max-size <KB>: Enforces an upper file size limit for compressed outputs.
Image Utility Toolbox Advanced Features
- Dynamic Inpainting & Bounding Boxes: Supports both seamless automated background restoration and manual pixel coordinate bounding boxes.
- Zero-Dependency Error Fallback: Automatically intercepts
ModuleNotFoundErrorduring agent runtimes, prompting automatic package fetching for frictionless execution. - ONNX-Accelerated Subject Extraction: Utilizes high-performance background removal to cleanly isolate portrait silhouettes without professional lighting setups.
- Automated Validation: Parses standard error streams and reports back clean diagnostic suggestions to user agents if file errors or script disruptions occur.
SKILL.md
Loading
Related Openclaw Skills
Skill Optimizer & Developer
A dedicated development and optimization assistant designed to automatically analyze, refine, and update existing AI agent skills and documentation.
Auto Rename Session Label
An internal OpenClaw hook that automatically titles new chat sessions using the active session's LLM model, with a robust first-message truncation fallback.
OpenAI-Compatible LLM Gateway
A keyless, drop-in OpenAI-compatible gateway that supports 17+ frontier AI models with pay-per-call USDC payments.
Tax BP Customer Service Knowledge Base (tax-bp-knowledge)
An intelligent customer service agent skill that dynamically pulls tax rules, FAQs, and invoicing workflows from a centralized knowledge base to automate business-line specific tax inquiries.
AI First Aid Guide
A specialized AI agent skill designed to deliver structured, life-saving first aid instructions and generate interactive HTML emergency response reports.
Tattoo Design Guide
An expert AI tattoo consultant skill that turns user preferences into comprehensive, interactive HTML visual design and safety reports.
Featured*
