Naver Search MCP Server

An MCP server that enables searching various content using the Naver Search API, analyzing data with Datalab, and utilizing Vision API capabilities.

Features

Search Features

• News Search
• Blog Search
• Shopping Search
• Image Search
• Knowledge-iN Search
• Book Search
• Encyclopedia Search
• Web Document Search
• Academic Article Search
• Cafe Post Search

Datalab Features

• Search Term Trend Analysis
• Shopping Category Trends
• Device-specific Shopping Trends (PC/Mobile)
• Shopping Trends by Gender/Age Groups
• Shopping Keyword Trends

Vision API Features

• Celebrity Face Detection
• Face Similarity Analysis

Installation and Execution

Environment Variables Setup

# Windows
set NAVER_CLIENT_ID=your_client_id
set NAVER_CLIENT_SECRET=your_client_secret

# Linux/Mac
export NAVER_CLIENT_ID=your_client_id
export NAVER_CLIENT_SECRET=your_client_secret

Run with NPX

npx @modelcontextprotocol/server-naver-search

Run with Docker

docker run -i --rm \
  -e NAVER_CLIENT_ID=your_client_id \
  -e NAVER_CLIENT_SECRET=your_client_secret \
  mcp/naver-search

Cursor Desktop Configuration

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "naver-search": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-naver-search"],
      "env": {
        "NAVER_CLIENT_ID": "your_client_id",
        "NAVER_CLIENT_SECRET": "your_client_secret"
      }
    }
  }
}

Or if using Docker:

{
  "mcpServers": {
    "naver-search": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "NAVER_CLIENT_ID=your_client_id",
        "-e",
        "NAVER_CLIENT_SECRET=your_client_secret",
        "mcp/naver-search"
      ]
    }
  }
}

Tools

Unified Search

• search
  • A unified search tool that can perform searches for all content types
  • Parameters:
    • type: Search type (news, blog, shop, etc.)
    • query: Search term
    • display: Number of results to display (default: 10)
    • start: Starting position for search results (default: 1)
    • sort: Sort method (sim: by similarity, date: by date)

Individual Search Tools

• search_news: News search
• search_blog: Blog search
• search_shop: Shopping search
• search_image: Image search
• search_kin: Knowledge-iN search
• search_book: Book search
• search_encyc: Encyclopedia search
• search_doc: Web document search
• search_academic: Academic article search
• search_cafe: Cafe post search

Each search tool accepts the following parameters:

• query: Search term
• display: Number of results to display (default: 10)
• start: Starting position for search results (default: 1)
• sort: Sort method (sim: by similarity, date: by date)

Datalab Tools

• datalab_search_trend

  • Analyze search term trends
  • Parameters:
    • startDate: Analysis start date (YYYY-MM-DD)
    • endDate: Analysis end date (YYYY-MM-DD)
    • timeUnit: Time unit for analysis (date, week, month)
    • keywords: List of keywords to analyze
    • category: Category to analyze within

• datalab_shopping_trend

  • Analyze shopping trends
  • Parameters:
    • startDate: Analysis start date (YYYY-MM-DD)
    • endDate: Analysis end date (YYYY-MM-DD)
    • timeUnit: Time unit for analysis (date, week, month)
    • category: Shopping category
    • device: Device type (pc, mobile, all)
    • gender: Gender filter (m, f, a)
    • ages: Age groups to analyze

Vision API Tools

• vision_face

  • Detect and analyze faces in images
  • Parameters:
    • image: Image URL or Base64 encoded image
    • mode: Analysis mode (face, celebrity)

• vision_face_compare

  • Compare faces for similarity
  • Parameters:
    • image1: First image URL or Base64 encoded image
    • image2: Second image URL or Base64 encoded image

Build

Docker build:

docker build -t mcp/naver-search .

License

MIT License