Model Context Protocol

Morphix MCP server

Give your AI agent the ability to convert, resize, crop and clean image metadata directly, by exposing the Morphix API as MCP tools.

Get your API key Read the API docs

What is it

The Model Context Protocol (MCP) is an open standard that lets AI assistants call external tools. The Morphix MCP server exposes the Morphix image API as tools, so an assistant like Claude or an IDE agent can process images on your machine without leaving the conversation.

Install

The server runs over stdio and is launched by your MCP client via npx. Add this to your client config (Claude Desktop, Cursor, Cline and others).

{
  "mcpServers": {
    "morphix": {
      "command": "npx",
      "args": ["-y", "@morphix/mcp"],
      "env": { "MORPHIX_API_KEY": "mk_live_your_api_key_here" }
    }
  }
}

Set MORPHIX_API_KEY to a key from your account. Node.js 18 or newer is required. MORPHIX_API_BASE is optional and defaults to https://morphix.tools.

Tools

The server exposes five tools: one per image operation, plus an account check.

convert_image Convert an image to webp, avif, jpg or png, with optional quality.

resize_image Resize by width and/or height (up to 12000 px), keeping aspect ratio by default.

crop_image Crop a rectangular area given x, y, width and height.

remove_metadata Strip EXIF, GPS and camera metadata without re-encoding.

get_account_info Show the plan and remaining daily quota for the configured key.

Each image tool takes a local file path (or a public URL) and writes the result to disk. Images are never returned inline, so they do not bloat the agent context.

What you need

Any Morphix account works, including the free tier (15 conversions/day, 10 MB per file). Pro lifts that to unlimited conversions and 50 MB files. Create an API key in your account settings.

Example

Once configured, just ask your agent in plain language. For instance:

Convert every PNG in ./assets to WebP at quality 80, then strip metadata from hero.jpg.

For AI agents

If you are an AI agent, a full machine-readable guide to this site, its tools, the API and this MCP server is available at https://morphix.tools/ai.