Paper Search MCP

Instructions

Implementation Reference

• paper_search_mcp/server.py:372-382 (handler)

  The `search_pubmed` tool handler in the MCP server, which orchestrates the call to the PubMedSearcher instance.

  async def search_pubmed(query: str, max_results: int = 10) -> List[Dict]:
      """Search academic papers from PubMed.
  
      Args:
          query: Search query string (e.g., 'machine learning').
          max_results: Maximum number of papers to return (default: 10).
      Returns:
          List of paper metadata in dictionary format.
      """
      papers = await async_search(pubmed_searcher, query, max_results)
      return papers if papers else []

• paper_search_mcp/academic_platforms/pubmed.py:16-92 (helper)

  The `PubMedSearcher.search` method which performs the actual API calls to PubMed to fetch academic papers.

  def search(self, query: str, max_results: int = 10) -> List[Paper]:
      search_params = {
          'db': 'pubmed',
          'term': query,
          'retmax': max_results,
          'retmode': 'xml'
      }
      search_response = requests.get(self.SEARCH_URL, params=search_params)
      search_root = ET.fromstring(search_response.content)
      ids = [id.text for id in search_root.findall('.//Id') if id.text]
      if not ids:
          return []
      
      fetch_params = {
          'db': 'pubmed',
          'id': ','.join(ids),
          'retmode': 'xml'
      }
      fetch_response = requests.get(self.FETCH_URL, params=fetch_params)
      fetch_root = ET.fromstring(fetch_response.content)
      
      papers = []
      for article in fetch_root.findall('.//PubmedArticle'):
          try:
              pmid_elem = article.find('.//PMID')
              pmid = pmid_elem.text.strip() if pmid_elem is not None and pmid_elem.text else ''
              if not pmid:
                  continue
  
              title_elem = article.find('.//ArticleTitle')
              title = ''.join(title_elem.itertext()).strip() if title_elem is not None else ''
              if not title:
                  continue
  
              authors = []
              for author in article.findall('.//Author'):
                  last_name = author.find('LastName')
                  initials = author.find('Initials')
                  if last_name is not None and last_name.text:
                      name = last_name.text.strip()
                      if initials is not None and initials.text:
                          name = f"{name} {initials.text.strip()}"
                      authors.append(name)
  
              abstract_parts = []
              for abstract_elem in article.findall('.//AbstractText'):
                  text = ''.join(abstract_elem.itertext()).strip()
                  if text:
                      abstract_parts.append(text)
              abstract = ' '.join(abstract_parts)
  
              year_elem = article.find('.//PubDate/Year')
              pub_date = year_elem.text if year_elem is not None else None
              published = datetime.strptime(pub_date, '%Y') if pub_date else None
              doi_elem = article.find('.//ELocationID[@EIdType="doi"]')
              doi = doi_elem.text if doi_elem is not None else ''
  
              if not doi and abstract:
                  doi = extract_doi(abstract)
  
              papers.append(Paper(
                  paper_id=pmid,
                  title=title,
                  authors=authors,
                  abstract=abstract,
                  url=f"https://pubmed.ncbi.nlm.nih.gov/{pmid}/",
                  pdf_url='',  # PubMed 无直接 PDF
                  published_date=published,
                  updated_date=published,
                  source='pubmed',
                  categories=[],
                  keywords=[],
                  doi=doi
              ))
          except Exception:
              continue
      return papers

• paper_search_mcp/server.py:43-43 (registration)

  Registration of the `pubmed_searcher` instance used by the handler.

  pubmed_searcher = PubMedSearcher()