Office Word MCP Server

Instructions

Implementation Reference

• word_document_server/tools/comment_tools.py:109-168 (handler)

  Main asynchronous handler implementing the tool logic: loads document, validates inputs, extracts and filters comments for the specified paragraph, includes paragraph text context, returns structured JSON response.

  async def get_comments_for_paragraph(filename: str, paragraph_index: int) -> str:
      """
      Extract comments for a specific paragraph in a Word document.
      
      Args:
          filename: Path to the Word document
          paragraph_index: Index of the paragraph (0-based)
          
      Returns:
          JSON string containing comments for the specified paragraph
      """
      filename = ensure_docx_extension(filename)
      
      if not os.path.exists(filename):
          return json.dumps({
              'success': False,
              'error': f'Document {filename} does not exist'
          }, indent=2)
      
      if paragraph_index < 0:
          return json.dumps({
              'success': False,
              'error': 'Paragraph index must be non-negative'
          }, indent=2)
      
      try:
          # Load the document
          doc = Document(filename)
          
          # Check if paragraph index is valid
          if paragraph_index >= len(doc.paragraphs):
              return json.dumps({
                  'success': False,
                  'error': f'Paragraph index {paragraph_index} is out of range. Document has {len(doc.paragraphs)} paragraphs.'
              }, indent=2)
          
          # Extract all comments
          all_comments = extract_all_comments(doc)
          
          # Filter for the specific paragraph
          from word_document_server.core.comments import get_comments_for_paragraph as core_get_comments_for_paragraph
          para_comments = core_get_comments_for_paragraph(all_comments, paragraph_index)
          
          # Get the paragraph text for context
          paragraph_text = doc.paragraphs[paragraph_index].text
          
          # Return results
          return json.dumps({
              'success': True,
              'paragraph_index': paragraph_index,
              'paragraph_text': paragraph_text,
              'comments': para_comments,
              'total_comments': len(para_comments)
          }, indent=2)
          
      except Exception as e:
          return json.dumps({
              'success': False,
              'error': f'Failed to extract comments: {str(e)}'
          }, indent=2)

• word_document_server/main.py:414-417 (registration)

  MCP tool registration via FastMCP @mcp.tool() decorator in the main server file, defining input parameters and docstring, delegating execution to the handler in comment_tools.

  def get_comments_for_paragraph(filename: str, paragraph_index: int):
      """Extract comments for a specific paragraph in a Word document."""
      return comment_tools.get_comments_for_paragraph(filename, paragraph_index)
  # New table column width tools

• word_document_server/core/comments.py:199-210 (helper)

  Core helper function that filters a list of comments to return only those associated with the given paragraph_index.

  def get_comments_for_paragraph(comments: List[Dict[str, Any]], paragraph_index: int) -> List[Dict[str, Any]]:
      """
      Get all comments for a specific paragraph.
      
      Args:
          comments: List of all comments
          paragraph_index: Index of the paragraph
          
      Returns:
          Comments for the specified paragraph
      """
      return [c for c in comments if c.get('paragraph_index') == paragraph_index]