Naver Search MCP Server

Instructions

Implementation Reference

• server.py:553-566 (handler)

  The handler function implementing the core logic of the 'search_doc' tool. It handles pagination, prepares API parameters, and invokes the shared _make_api_call to retrieve and format results from the Naver 'doc.json' endpoint.
  async def search_doc(query: str, display: int = 10, page: int = 1) -> str: """ Searches for academic papers, reports, etc. using the given keyword. The page parameter allows for page navigation. Args: query (str): The keyword to search for display (int, optional): The number of results to display. Default is 10. page (int, optional): The starting page number. Default is 1. """ start = calculate_start(page, display) display = min(display, 100) params = {"query": query, "display": display, "start": start} return await _make_api_call("doc.json", params, DocResult, "Academic Papers")
• server.py:549-552 (registration)

  The @mcp.tool decorator registering the 'search_doc' tool, specifying its name and description for the MCP protocol.
  @mcp.tool( name="search_doc", description="Searches for academic papers, reports, etc. using the given keyword. The page parameter allows for page navigation." )
• server.py:128-128 (schema)

  Pydantic model (DocResult) used for schema validation of the API response data specific to document/academic papers search, which is parsed and formatted in the handler.
  class DocResult(SearchResultBase): items: List[DocItem]