RSpace MCP Server

Instructions

Implementation Reference

• main.py:355-404 (handler)

  The handler function for the 'find_documents_by_content' MCP tool. It uses RSpace's AdvancedQueryBuilder for full-text search on document content with support for AND/OR operators and post-query exclusion filtering on document metadata.

  @mcp.tool(tags={"rspace", "search"})
  def find_documents_by_content(
      content_terms: List[str],
      operator: Literal["and", "or"] = "and",
      exclude_terms: List[str] = None,
      order_by: str = "lastModified desc",
      page_size: int = 20
  ) -> dict:
      """
      Advanced content-based document search
      
      Usage: Find documents containing specific content terms
      
      Parameters:
      - content_terms: List of terms that should appear in document content
      - operator: "and" (all terms must appear) or "or" (any term can appear)
      - exclude_terms: Optional list of terms to exclude from results
      - order_by: Sort results by field
      - page_size: Number of results to return
      
      Returns: Dictionary with search results
      
      Example: find_documents_by_content(["DNA", "extraction"], operator="and")
      """
      builder = AdvancedQueryBuilder(operator=operator)
      
      for term in content_terms:
          builder.add_term(term, AdvancedQueryBuilder.QueryType.FULL_TEXT)
      
      # Note: RSpace API doesn't directly support exclusion, but we can filter results
      advanced_query = builder.get_advanced_query()
      results = eln_cli.get_documents_advanced_query(
          advanced_query=advanced_query,
          order_by=order_by,
          page_number=0,
          page_size=page_size
      )
      
      # Filter out documents containing excluded terms if specified
      if exclude_terms and 'documents' in results:
          filtered_docs = []
          for doc in results['documents']:
              # Check if any exclude terms are in the document name or other available text
              doc_text = (doc.get('name', '') + ' ' + doc.get('tags', '')).lower()
              if not any(exclude_term.lower() in doc_text for exclude_term in exclude_terms):
                  filtered_docs.append(doc)
          results['documents'] = filtered_docs
          results['totalHits'] = len(filtered_docs)
      
      return results