Paper Search MCP

Instructions

Implementation Reference

• paper_search_mcp/server.py:1146-1156 (handler)

  The MCP tool registration and entry point for 'download_doaj'.

  @mcp.tool()
  async def download_doaj(paper_id: str, save_path: str = "./downloads") -> str:
      """Download PDF for a paper from DOAJ.
  
      Args:
          paper_id: DOAJ paper identifier.
          save_path: Directory to save the PDF (default: './downloads').
      Returns:
          str: Path to downloaded PDF.
      """
      return doaj_searcher.download_pdf(paper_id, save_path)

• paper_search_mcp/academic_platforms/doaj.py:360-405 (handler)

  The actual logic implementation for downloading a DOAJ PDF within the DOAJSearcher class. (Snippet simplified)

  def download_pdf(self, paper_id: str, save_path: str) -> str:
      """Download PDF for a DOAJ article.
  
      DOAJ provides direct PDF links for open access articles.
  
      Args:
          paper_id: DOAJ article ID or DOI
          save_path: Directory to save PDF
  
      Returns:
          Path to saved PDF file
  
      Raises:
          ValueError: If paper not found or no PDF available
          IOError: If download fails
      """
      # Try to get paper info first
      papers = self.search(paper_id, max_results=1)
      if not papers:
          raise ValueError(f"DOAJ article not found: {paper_id}")
  
      paper = papers[0]
      if not paper.pdf_url:
          # Try to construct PDF URL from DOI
          if paper.doi:
              # Some publishers provide direct PDF links via DOI
              pdf_url = f"https://doi.org/{paper.doi}"
              # But we need to check if it's actually a PDF
              # For now, try the URL
              paper.pdf_url = pdf_url
          else:
              raise ValueError(f"No PDF available for DOAJ article: {paper_id}")
  
      # Download PDF
      import os
      response = self.session.get(paper.pdf_url, timeout=30)
      response.raise_for_status()
  
      # Check if response is actually PDF
      content_type = response.headers.get('content-type', '')
      if 'pdf' not in content_type.lower() and not paper.pdf_url.lower().endswith('.pdf'):
          logger.warning(f"Response may not be PDF: {content_type}")
  
      os.makedirs(save_path, exist_ok=True)
  
      # Create safe filename