# Watermark & Resize Studio MCP [Health: Active]

**Category:** 🎥 Multimedia Process  
**Repository:** https://github.com/Miladmet/watermark-studio-mcp  
**GitHub Stars:** 0  
**npm Downloads (last month):** 714  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/watermark-studio-mcp

## Description
Batch image resizing, watermarking, EXIF stripping, OpenGraph cards and favicon generation. Local stdio server.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "watermark-resize-studio-mcp": {
    "command": "npx",
    "args": ["-y","watermark-studio-mcp"]
  }
}
```

## Documentation

## What the Watermark & Resize Studio MCP MCP server does

Watermark & Resize Studio MCP MCP server gives MCP-compatible assistants tools for working with image files stored on the local machine. Its image-processing functions cover resizing, format conversion, optimization, watermark composition, and photo metadata removal. The server also generates OpenGraph and Twitter preview cards, plus a favicon and progressive web app icon set from a source logo.

The server fits the 🎥 Multimedia Process category. It is relevant when an agent needs to prepare product images for online platforms, create branded assets, or sanitize photos before publication. Processing is local and uses the Sharp image library with the libvips engine; the README states that files are not uploaded to a cloud service.

## How it works

The server runs as a local stdio process. Each tool accepts paths to source files or folders and writes the result to an output path or a documented default location. `resize_for_platform` supports presets for Shopify, Etsy, Instagram formats, YouTube thumbnails, Facebook posts, and Pinterest pins, as well as custom dimensions. It can produce WebP, JPEG, PNG, or the original format, with configurable quality and fit behavior.

`apply_watermark` composites either text or a PNG/SVG logo. It supports six placement modes, including tiled output, along with opacity, color, font size, and margins. `strip_photo_metadata` removes EXIF, GPS, camera serial, and XMP edit-history data, with optional output-format conversion.

For directory workflows, `batch_process_folder` processes images in parallel. It can combine platform resizing, format conversion, text watermarking, and metadata stripping, with a configurable concurrency limit. `generate_social_card` creates a 1200x630 card with a selectable gradient theme, title, optional subtitle, brand label, logo, or background image. `generate_favicon_ico_pack` creates browser, Apple, Android, and manifest files from one logo image.

## Setup and configuration

Watermark & Resize Studio MCP MCP server requires Node.js 18 or newer. The package is available through npm, so a build step is not described. A local client can launch it with `npx -y watermark-studio-mcp`.

For Claude Desktop, add an `mcpServers` entry using that command to the client configuration file. On Windows, the README identifies `%APPDATA%\\Claude\\claude_desktop_config.json`; on macOS, it identifies `~/Library/Application Support/Claude/claude_desktop_config.json`. Cursor uses the same command in `.cursor/mcp.json`. The README also identifies Cline as a supported assistant, but does not provide a Cline-specific configuration example.

## Tools and capabilities

- Resize images with platform presets or custom dimensions.
- Convert and optimize images as WebP, JPEG, or PNG.
- Add text, logo, positioned, or tiled watermarks.
- Remove location and creator-related photo metadata.
- Process folders with parallel workers.
- Generate social preview cards and favicon/PWA assets.

## Limitations and notes

The tools operate on local paths, so the assistant and server need access to the relevant files and destination directories. Watermarking requires at least one of `text` or `logo_path`. The README documents supported input and output options, but does not describe a remote service, authentication flow, environment variables, or an HTTP transport. The stated compression reduction of 85–95% is presented as a project feature and may depend on the input image, selected format, and quality setting.

_Full upstream README: https://allmcps.com/mcp/watermark-studio-mcp/readme_