HeyBeauty

HeyBeauty MCP Server for Virtual TryOn

javascript

18

GitHub Stars

javascript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "chatmcp-heybeauty-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "heybeauty-mcp"
      ],
      "env": {
        "HEYBEAUTY_API_KEY": "your_heybeauty_api_key"
      }
    }
  }
}

You can run the HeyBeauty MCP Server to perform virtual try-ons by submitting cloth selections and user images, then retrieving task progress and results. This server exposes clothes as resources, supports task submission and querying, and provides prompts to guide large language models in generating try-on results.

How to use

To use the HeyBeauty MCP Server, start the local MCP server process and connect with an MCP client. You can browse and access clothes via cloth:// URIs, submit a try-on task with user and cloth images, and then query the status and details of that task. Use the provided prompts to structure LLm-assisted try-ons and receive organized results.

How to install

Prerequisites: Node.js installed on your system, plus an active HeyBeauty API key.

Install dependencies for the server.

Build the server for production or development use.

Additional content

Configuration notes provide how to run the MCP server with your client and how to integrate with Claude Desktop if needed. You’ll also find debugging guidance for MCP servers that communicate over stdio.

Available tools

submit_tryon_task

Submit a try-on task by providing the user image URL, cloth image URL, cloth id, and cloth description. The server stores the task in its state and returns a task identifier.

query_tryon_task

Query a try-on task by task id to retrieve the current status and details of the try-on operation.

tryon_cloth

Provide a prompt for LLm-based try-on that combines user image URL, cloth image URL, cloth id, and cloth description into a structured prompt for generation.