ResearchMCP

Instructions

Implementation Reference

• src/server.py:209-270 (handler)

  Main execution logic for the find_research_gaps tool: searches papers using PaperFetcher, fetches abstracts, and formats comprehensive instructions for LLM-based gap analysis.

  @mcp.tool()
  def find_research_gaps(query: str, num_papers: int = 5) -> str:
      """
      Analyze multiple papers on a topic to identify research gaps and unanswered questions.
  
      Args:
          query: Research topic to analyze
          num_papers: Number of papers to analyze (default: 5, max: 10)
  
      Returns:
          Analysis of research gaps, limitations, and future research directions
      """
      if num_papers > 10:
          num_papers = 10
  
      papers = fetcher.search_papers(query=query, max_results=num_papers, sort_by="cited_by_count")
  
      if papers and "error" in papers[0]:
          return papers[0]["error"]
  
      if not papers:
          return f"No papers found for query: {query}"
  
      result = f"**Research Gap Analysis for: '{query}'**\n"
      result += f"**Analyzing {len(papers)} highly-cited papers**\n\n"
  
      result += "**Papers Analyzed:**\n"
      paper_ids = []
      for i, paper in enumerate(papers, 1):
          result += f"{i}. {paper['title']} ({paper['publication_year']}) - {paper['cited_by_count']} citations\n"
          paper_ids.append(paper['id'])
  
      result += "\n**Fetching abstracts for deep analysis...**\n\n"
  
      abstracts_data = []
      for i, paper_id in enumerate(paper_ids, 1):
          paper_detail = fetcher.fetch_paper_by_id(paper_id)
          if "error" not in paper_detail:
              abstract_text = fetcher.get_paper_abstract(paper_detail)
              abstracts_data.append(f"**Paper {i}:** {paper_detail['title']}\n{abstract_text}\n")
  
      result += "".join(abstracts_data)
  
      result += "\n**Gap Analysis Instructions:**\n"
      result += "Based on the abstracts above, please identify:\n\n"
      result += "1. **Unanswered Research Questions:**\n"
      result += "   - What questions do these papers raise but not answer?\n"
      result += "   - What do the authors suggest for future research?\n\n"
      result += "2. **Methodological Limitations:**\n"
      result += "   - What limitations do the authors acknowledge?\n"
      result += "   - What methods or approaches are missing?\n\n"
      result += "3. **Understudied Areas:**\n"
      result += "   - What aspects of the topic receive less attention?\n"
      result += "   - What populations, contexts, or scenarios are not covered?\n\n"
      result += "4. **Contradictions & Inconsistencies:**\n"
      result += "   - Where do findings conflict?\n"
      result += "   - What requires further investigation to resolve?\n\n"
      result += "5. **Emerging Opportunities:**\n"
      result += "   - What new research directions are suggested?\n"
      result += "   - What interdisciplinary connections could be made?\n"
  
      return result

• src/server.py:210-220 (schema)

  Type hints and docstring defining input parameters (query: str, num_papers: int=5) and output (str) for the tool.

  def find_research_gaps(query: str, num_papers: int = 5) -> str:
      """
      Analyze multiple papers on a topic to identify research gaps and unanswered questions.
  
      Args:
          query: Research topic to analyze
          num_papers: Number of papers to analyze (default: 5, max: 10)
  
      Returns:
          Analysis of research gaps, limitations, and future research directions
      """

• src/server.py:209-209 (registration)

  Decorator that registers the find_research_gaps function as an MCP tool.

  @mcp.tool()