Gemini Watermark Remover
About
Gemini Watermark Remover is an MCP server that removes Gemini AI watermarks from generated images. It provides a single tool, geminiremovewatermark, for use in MCP-compatible clients like Claude Desktop and OpenCode.
Explore
- Removes Gemini AI watermark from images
- Supports PNG, JPG, and WEBP input formats
- Configurable output quality and compression
- Output can be markdown or JSON
- Runs locally via stdio transport
- Built on the gemini-watermark-remover library
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Gemini Watermark RemoverCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
- Node.js >= 22
- npm
npm install
For debugging after modifying source code, you can use the locally built version:
Claude Desktop:
``json``
{
"mcpServers": {
"gemini-watermark-remove": {
"command": "node",
"args": ["/path/to/gemini-watermark-remover-mcp/dist/index.js"]
}
}
}
gemini_remove_watermark
Removes the Gemini AI watermark from generated images. Args: - input_path: Path to the input image file (PNG, JPG, WEBP) - output_path: Path where the processed image will be saved (PNG, JPG, WEBP) - response_format: 'markdown' or 'json' (default: 'markdown') - jpeg_quality: JPEG output quality 1-100 (default: 95) - webp_quality: WEBP output quality 1-100 (default: 80) - png_compression_level: PNG compression level 0-9 (default: 6) Output formats: - PNG: Configurable compression (0-9) - JPG: Configurable quality (1-100) - WEBP: Configurable quality (1-100)
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"gemini watermark remover": {
"gemini-watermark-remove": {
"command": "npx",
"args": [
"-y",
"gemini-watermark-remover-mcp"
]
}
}
}
}McpServers
{
"gemini-watermark-remove": {
"command": "npx",
"args": [
"-y",
"gemini-watermark-remover-mcp"
]
}
}Gemini Watermark Remove MCP Server
MCP server for removing Gemini AI watermarks from generated images.Local Stdio Mode
Claude Desktop
``json
{
"mcpServers": {
"gemini-watermark-remove": {
"command": "npx",
"args": ["-y", "gemini-watermark-remover-mcp"]
}
}
}
`
OpenCode
`json
{
"mcp": {
"gemini-watermark-remove": {
"command": ["npx", "-y", "gemini-watermark-remover-mcp"],
"type": "local",
"enabled": true
}
}
}
`
Development
Prerequisites
- Node.js >= 22
- npm
Installation
`bash
Clone or download the project
cd gemini-watermark-remover-mcp
Install dependencies
npm install
Build the project
npm run build
`
Running
Development mode (hot reload):
`bash
npm run dev
`
Production mode:
`bash
npm start
or
node dist/index.js
`
Local Development Debug Config
For debugging after modifying source code, you can use the locally built version:
Claude Desktop:
`json
{
"mcpServers": {
"gemini-watermark-remove": {
"command": "node",
"args": ["/path/to/gemini-watermark-remover-mcp/dist/index.js"]
}
}
}
``
MCP Tool Usage
gemini_remove_watermark
Removes the Gemini AI watermark from generated images. Parameters: | Parameter | Type | Required | Default | Description | |-----------|------|----------|---------|-------------| | input_path | string | Yes | - | Input image path (PNG, JPG, WEBP) | | output_path | string | Yes | - | Output image path | | response_format | string | No | markdown | Output format: markdown or json | | jpeg_quality | number | No | 95 | JPEG quality (1-100) | | webp_quality | number | No | 80 | WEBP quality (1-100) | | png_compression_level | number | No | 6 | PNG compression level (0-9) |License
This project is licensed under the MIT LicenseAcknowledgments
This project is based on journey-ad/gemini-watermark-remover. Thanks for the original implementation!Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
