Paper Search MCP

Instructions

Implementation Reference

• paper_search_mcp/server.py:521-534 (handler)

  The handler for the read_arxiv_paper tool in the MCP server, which delegates to the ArxivSearcher.

  async def read_arxiv_paper(paper_id: str, save_path: str = "./downloads") -> str:
      """Read and extract text content from an arXiv paper PDF.
  
      Args:
          paper_id: arXiv paper ID (e.g., '2106.12345').
          save_path: Directory where the PDF is/will be saved (default: './downloads').
      Returns:
          str: The extracted text content of the paper.
      """
      try:
          return arxiv_searcher.read_paper(paper_id, save_path)
      except Exception as e:
          print(f"Error reading paper {paper_id}: {e}")
          return ""

• paper_search_mcp/academic_platforms/arxiv.py:90-117 (handler)

  The actual implementation logic that reads a PDF file and extracts its text content.

  def read_paper(self, paper_id: str, save_path: str = "./downloads") -> str:
      """Read a paper and convert it to text format.
      
      Args:
          paper_id: arXiv paper ID
          save_path: Directory where the PDF is/will be saved
          
      Returns:
          str: The extracted text content of the paper
      """
      # First ensure we have the PDF
      pdf_path = f"{save_path}/{paper_id}.pdf"
      if not os.path.exists(pdf_path):
          pdf_path = self.download_pdf(paper_id, save_path)
      
      # Read the PDF
      try:
          reader = PdfReader(pdf_path)
          text = ""
          
          # Extract text from each page
          for page in reader.pages:
              text += page.extract_text() + "\n"
          
          return text.strip()
      except Exception as e:
          print(f"Error reading PDF for paper {paper_id}: {e}")
          return ""

• paper_search_mcp/server.py:520-521 (registration)

  The MCP tool registration decorator for the read_arxiv_paper function.

  @mcp.tool()
  async def read_arxiv_paper(paper_id: str, save_path: str = "./downloads") -> str: