Server & API/Model Context Protocol (MCP)
Docs/Server & API/Model Context Protocol (MCP)

Model Context Protocol (MCP)

Connect Imagerry to Claude Desktop, Cursor, and Zed AI assistants.

The imagerry mcp command exposes image styling tools and presets to AI assistants using the open Model Context Protocol (MCP).

Start the MCP Server

|bash
imagerry mcp

Supported Tools & Capabilities

The Imagerry MCP server registers the following native tool:

  • `customize_image`: Applies custom styles, presets, gradients, padding, and formatting directly to local image files on the user's computer.
  • inputPath (string, required): Absolute file path to the input image.
  • outputPath (string, required): Absolute file path to save the styled output image.
  • preset (string, optional): Built-in preset (default, mesh, gradient), path to a custom JSON preset file, or a raw JSON template string.
  • mode (string, optional): Workspace tool mode (defaults to image-customizer).

Claude Desktop Setup

Add the following entry to your claude_desktop_config.json:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
|json
{
  "mcpServers": {
    "imagerry": {
      "command": "imagerry",
      "args": ["mcp"],
      "env": {
        "IMAGERRY_LICENSE_KEY": "YOUR_PRO_LICENSE_KEY"
      }
    }
  }
}

Cursor IDE Setup

In Cursor Settings, Features, MCP Servers, Add New Server:

  • Name: imagerry
  • Type: command
  • Command: imagerry mcp
  • Environment Variables: IMAGERRY_LICENSE_KEY=YOUR_KEY
Note

Clean stdio Protocol

Imagerry MCP strictly adheres to JSON-RPC stdio protocol standards. All logs and diagnostics are directed to stderr, ensuring reliable, error-free communication with AI clients.

Found an issue or typo? Report on GitHub
Imagerry CLI Docs • v0.1.6-beta