wavespeed-watermark-remover
Remove watermarks, logos, captions, and text overlays from images and videos using WaveSpeed AI. Intelligently detects and removes watermarks while preserving texture and background. Supports images and videos up to 10 minutes. Use when the user wants to remove watermarks or text overlays from media.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/chengzeyi/wavespeed-watermark-removerWhat This Skill Does
The wavespeed-watermark-remover skill provides an advanced interface for the WaveSpeed AI engine, specialized in removing unwanted visual elements from media. Whether it is an intrusive watermark on a stock photo, a persistent logo on a YouTube clip, or distracting text overlays on social media assets, this skill uses temporal-aware inpainting and sophisticated texture reconstruction to restore your media. By processing frames intelligently, it ensures that the background texture, color grading, and lighting remain consistent with the original file, effectively masking the area where the watermark previously resided.
Installation
To add this skill to your OpenClaw environment, use the provided CLI command. Ensure you have your WaveSpeed API key ready before initializing the skill configuration.
Command: clawhub install openclaw/skills/skills/chengzeyi/wavespeed-watermark-remover
Once installed, you must export your API key to your environment variables using export WAVESPEED_API_KEY="your-api-key" to authenticate requests properly. You can generate your unique key via the WaveSpeed developer portal.
Use Cases
- Content Creation: Quickly clean up stock footage or promotional images before including them in your own design projects.
- Content Archiving: Remove burned-in captions or dates from old videos to preserve the aesthetic quality of historical footage.
- Workflow Automation: Integrate watermark removal into a batch processing pipeline, ensuring all imported assets meet your branding standards automatically.
- Social Media Management: Edit and repurpose trending videos or viral images by removing original watermarks before reposting.
Example Prompts
- "OpenClaw, please remove the watermark from this image: [image_url] and save it as a PNG."
- "I have a video at [video_url] that has a distracting logo in the bottom right corner. Can you use WaveSpeed to clean it up?"
- "Process this file [file_path] and strip out all text overlays; return the clean version for my presentation."
Tips & Limitations
For the best results, ensure the media provided is of high resolution. While the AI is robust, very complex backgrounds or watermarks covering large portions of faces may require manual touch-ups. For video, remember that the tool is limited to 10-minute clips; longer videos should be split into segments before processing. Always check your output format, particularly for images, to ensure the best color reproduction.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-chengzeyi-wavespeed-watermark-remover": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: external-api, file-read
Related Skills
wavespeed-face-swapper
Swap faces in images and videos using WaveSpeed AI. Supports image face swap and video face swap with multi-face targeting. Produces watermark-free results with automatic lighting and skin tone adaptation. Use when the user wants to replace a face in an image or video with another face.
wavespeed-infinitetalk
Generate talking head videos from a portrait image and audio using WaveSpeed AI's InfiniteTalk model. Produces lip-synced video up to 10 minutes long at 480p or 720p. Supports optional mask images to target specific faces and text prompts for additional guidance. Use when the user wants to animate a face with audio or create talking avatar videos.
wavespeed-minimax-speech-26
Convert text to speech using MiniMax Speech 2.6 Turbo via WaveSpeed AI. Features ultra-human voice cloning, sub-250ms latency, 40+ languages, emotion control, and 200+ voice presets. Use when the user wants to generate speech audio from text.
wavespeed-seedream-45
Generate and edit images using ByteDance's Seedream V4.5 model via WaveSpeed AI. Supports text-to-image generation and multi-image editing with custom resolutions up to 4096x4096. Features enhanced typography for posters and logos. Use when the user wants to create or edit images with high-quality text rendering.
wavespeed-nano-banana-2
Generate and edit images using Google's Nano Banana 2 model via WaveSpeed AI. Supports text-to-image generation and image editing with natural language prompts. Features native 4K resolution, flexible aspect ratios including ultra-narrow (1:8, 8:1), multilingual text rendering, and camera-style controls. Use when the user wants to create images from text or edit existing images.