ResearchMCP

Instructions

Implementation Reference

• src/server.py:9-44 (handler)

  The main MCP tool handler for 'search_papers', registered via @mcp.tool(). Handles input parameters, calls the PaperFetcher helper to search OpenAlex API, formats results into a markdown string with paper details.
  @mcp.tool() def search_papers(query: str, max_results: int = 5, year_from: Optional[int] = None) -> str: """ Search for academic papers on OpenAlex based on a query. Args: query: The research topic or keywords to search for max_results: Maximum number of papers to return (default: 5) year_from: Only include papers from this year onwards (optional) Returns: A formatted string with paper details including titles, authors, citations, and URLs """ papers = fetcher.search_papers( query=query, max_results=max_results, year_from=year_from ) if papers and "error" in papers[0]: return papers[0]["error"] if not papers: return f"No papers found for query: {query}" result = f"Found {len(papers)} papers for '{query}':\n\n" for i, paper in enumerate(papers, 1): result += f"{i}. **{paper['title']}**\n" result += f" Authors: {paper['authors']}\n" result += f" Year: {paper['publication_year']}\n" result += f" Citations: {paper['cited_by_count']}\n" result += f" URL: {paper['url']}\n" result += f" ID: {paper['id']}\n\n" return result
• src/tools/paper_fetcher.py:13-57 (helper)

  Supporting utility method in PaperFetcher class that implements the core logic: constructs OpenAlex API request for paper search, handles HTTP calls, parses responses into standardized paper dictionaries using _parse_paper.
  def search_papers( self, query: str, max_results: int = 5, sort_by: str = "cited_by_count", year_from: Optional[int] = None ) -> List[Dict]: """ Search for papers on OpenAlex. Args: query: Search keywords max_results: Maximum number of results sort_by: Sort by 'cited_by_count', 'publication_date', or 'relevance' year_from: Only papers from this year onwards Returns: List of paper dictionaries with title, authors, abstract, etc. """ url = f"{self.BASE_URL}/works" params = { "search": query, "per_page": max_results, "sort": sort_by + ":desc" } if year_from: params["filter"] = f"publication_year:>{year_from-1}" try: response = self.client.get(url, params=params) response.raise_for_status() data = response.json() papers = [] for work in data.get("results", []): paper = self._parse_paper(work) papers.append(paper) return papers except httpx.HTTPError as e: return [{"error": f"Failed to fetch papers: {str(e)}"}]