mcp-pdf-tools

Instructions

Implementation Reference

• src/pdf_tools/server.py:152-182 (handler)

  Executes the merge-pdfs tool by merging the provided PDF files using PyPDF2.PdfMerger and saving to the output path, with error handling.

  if name == "merge-pdfs":
      input_paths = arguments.get("input_paths", [])
      output_path = arguments.get("output_path")
      
      if not input_paths or not output_path:
          raise ValueError("Missing required arguments")
  
      merger = PyPDF2.PdfMerger()
      
      try:
          # Add each PDF to the merger
          for path in input_paths:
              with open(path, 'rb') as pdf_file:
                  merger.append(pdf_file)
          
          # Write the merged PDF
          with open(output_path, 'wb') as output_file:
              merger.write(output_file)
          
          return [types.TextContent(
              type="text",
              text=f"Successfully merged {len(input_paths)} PDFs into {output_path}"
          )]
          
      except Exception as e:
          return [types.TextContent(
              type="text",
              text=f"Error merging PDFs: {str(e)}"
          )]
      finally:
          merger.close()

• src/pdf_tools/server.py:21-39 (registration)

  Registers the merge-pdfs tool with the MCP server, providing name, description, and input schema validation.

  types.Tool(
      name="merge-pdfs",
      description="Merge multiple PDF files into a single PDF",
      inputSchema={
          "type": "object",
          "properties": {
              "input_paths": {
                  "type": "array",
                  "items": {"type": "string"},
                  "description": "List of input PDF file paths"
              },
              "output_path": {
                  "type": "string",
                  "description": "Output path for merged PDF"
              }
          },
          "required": ["input_paths", "output_path"]
      }
  ),