home / mcp / k-beauty mcp server

K-Beauty MCP Server

Provides real-time K-Beauty information, AI skin analysis, and personalized routines via MCP.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "alexai-mcp-k-beauty-mcp": {
      "command": "python3",
      "args": [
        "kbeauty_mcp.py"
      ]
    }
  }
}

You run an MCP server that provides real-time K-Beauty information, AI-powered skin analysis, and personalized routines through an extensible client experience. This server lets you query brands, products, ingredients, and tailored skincare plans, while also supporting AI-driven skin analysis and live web search for up-to-date trends and pricing.

How to use

You connect to the K-Beauty MCP Server with an MCP client. Start by launching the local server from your development environment, then point your client to the running process. Use the available tools to explore brands, retrieve product details, analyze ingredients, get personalized routines, compare products, and trigger AI skin analysis from a selfie (when you have an image). The workflow is designed for practical, end-user actions such as discovering affordable brands, researching specific products, building a skin-type specific routine, and analyzing skin concerns to match targeted products.

Key capabilities you can use through the MCP client include: - Search K-Beauty brands and brands histories - Get detailed product information plus pricing and reviews - Analyze ingredients and understand safety and benefits - Generate personalized skincare routines based on skin type and concerns - Compare multiple products side-by-side - Analyze skin from a selfie and receive zone-by-zone analysis plus routine recommendations (when available)

How to install

Prerequisites: - Python 3.8+ installed on your system - Git installed to clone repositories - Internet access for initial dependency installation (or offline mode if you have a curated data package)

Step-by-step installation and startup: 1. Clone the MCP server repository to your working directory 2. Change into the project directory 3. Install dependencies from the requirements file 4. Run the MCP server script to start the local service

# Clone the MCP server repository
git clone <your-repo-url> k-beauty-mcp
cd k-beauty-mcp

# Install dependencies
pip install -r requirements.txt

# Start the MCP server (adjust path to the actual script if needed)
python3 kbeauty_mcp.py

Claude Desktop configuration

To connect Claude Desktop to your local MCP server, add a config entry that points to the server script and working directory.

{
  "mcpServers": {
    "kbeauty": {
      "command": "python3",
      "args": ["kbeauty_mcp.py"],
      "cwd": "."
    }
  }
}

Quick test (local server)

After starting the server, perform simple queries through your MCP client to verify connectivity. Try questions like “Tell me about COSRX snail essence” or “Recommend a K-Beauty routine for sensitive skin” to confirm responsive results and route through the available tools.

Available tools

search_kbeauty_brands

Search and get information about K-Beauty brands, including brand details, history, and popular products.

get_product_info

Get detailed product information such as benefits, pricing, and reviews for a given brand and product.

analyze_ingredients

Analyze skincare ingredients to explain benefits, safety, and usage recommendations.

recommend_routine

Generate a personalized skincare routine based on skin type, concerns, and routine type.

compare_products

Provide a side-by-side comparison of multiple K-Beauty products highlighting pros and cons.

analyze_skin_photo

AI-powered facial skin analysis from an uploaded image, yielding zone-by-zone results and a tailored routine.

skin_concern_matcher

Match specific skin concerns to targeted products, considering skin type and budget.