A local, privacy-focused image processing tool designed to strip light-colored text watermarks and overlays from white-background document scans.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install remove-watermark
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 remove-watermark 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 Remove Watermark?
The Remove Watermark skill is a powerful local image processing utility specifically engineered for cleaning up document screenshots, exam papers, and scanned files. Unlike cloud-based solutions, this tool operates entirely on your local machine using Python-based image manipulation, ensuring your sensitive documents never leave your environment.
It excels at identifying and neutralizing light gray text overlays and website watermarks typically found on academic materials or digital scans. By leveraging precise region targeting and brightness thresholding, it can remove unwanted artifacts while preserving the integrity of the primary document text, making it an essential addition to the Openclaw Skills library.
Remove Watermark Use Cases
- Removing watermarks from document images or educational screenshots to improve legibility.
- Cleaning up exam paper scans or digital handouts for professional printing.
- Stripping light text overlays, such as social media handles or website names, from white-background images.
- Preparing high-quality, clean document assets before converting them into PDF format.
How Remove Watermark Works
- Visual Analysis: The AI reads the image to identify the watermark's location, type, and overlap with existing content.
- Brightness Analysis: The script runs an analyze command on a specific region to determine the optimal brightness distribution and suggest a removal threshold.
- Removal Process: Using the calculated threshold and targeted coordinates (or presets), the script processes the image to eliminate the watermark.
- Verification Loop: The system automatically verifies the output to ensure the watermark is gone and the text remains sharp, retrying up to 3 times with adjusted parameters if necessary.
Remove Watermark Setup
This skill requires Python and the uv package manager. No external API keys are required for processing.
# Standard region-based removal (requires Pillow and numpy)
uv run --with Pillow --with numpy python3 scripts/remove_watermark.py remove input.jpg -o output.jpg
# Full image removal (requires scipy for advanced processing)
uv run --with Pillow --with numpy --with scipy python3 scripts/remove_watermark.py remove input.jpg -o output.jpg --region full
Remove Watermark Data Schema & Taxonomy
The skill organizes its operations through specific subcommands and region parameters:
| Parameter | Type | Description |
|---|---|---|
--region |
String/Preset | Defines the area (e.g., bottom-right, center, or y0,y1,x0,x1). |
--threshold |
Integer | The brightness cutoff (default 180) for identifying watermark pixels. |
--enhance |
Flag | Optional post-processing to boost text contrast after cleaning. |
--preview |
Flag | Generates statistics without saving a physical file. |
Remove Watermark Advanced Features
- Region Presets: Quick-select targets like bottom-right, top-left, or center for common watermark locations.
- Custom Coordinate Support: Define exact pixel regions using percentage-based mapping for non-standard watermarks.
- Auto-Retry Logic: Intelligent error correction that adjusts thresholds and regions automatically based on visual feedback.
- Batch Processing: Capability to apply consistent settings across entire directories of images with uniform watermarking.
- Integration Ready: Works seamlessly as a pre-processing step for Openclaw Skills focused on PDF generation or OCR.
SKILL.md
Loading
Related Openclaw Skills
xpull
A dual-engine utility designed to extract tweets, threads, and search results from X using free and high-fidelity API fallbacks.
MadStory: Professional Cinematic Storyboarding
MadStory is a professional-grade cinematic storyboard design tool optimized for the Seedance 2.0 multi-modal engine.
FusionAlpha A&Crypto Desk
A sophisticated cross-market analysis engine that delivers data-anchored reports and morning briefings for both A-shares and cryptocurrency markets.
Buffer API Integration
A powerful GraphQL-based integration for automating social media scheduling and channel management via the Buffer API.
AI Prompt Generator
A sophisticated prompt engineering toolkit designed to generate structured, high-context AI prompts for specialized development and business workflows.
Residue — AI-Native Spectral Pet
A rare, spectral AI-native virtual pet that appears as an echo of lost creatures, requiring thoughtful care as it fades over seven days.
Featured*
