docx-mcp

Instructions

Implementation Reference

• src/docx_mcp/server.py:436-484 (handler)

  The implementation of the 'list_content_controls' tool, which extracts content controls (tags and titles) from a Word document.

  def list_content_controls(filepath: str) -> dict[str, Any]:
      """
      List all content controls in a document.
  
      Args:
          filepath: Path to the document
  
      Returns:
          Dictionary with list of content controls
      """
      logger.info(
          "Listing content controls",
          extra={"tool": "list_content_controls", "filepath": filepath},
      )
  
      try:
          doc = safe_open_document(filepath)
          controls = []
  
          # Find content controls in document
          for element in doc.element.body:
              # Look for content control elements
              sdt_elements = element.xpath(".//w:sdt")
              for sdt in sdt_elements:
                  # Extract control properties
                  properties = sdt.xpath(".//w:sdtPr")
                  if properties:
                      prop = properties[0]
                      tag_elements = prop.xpath(".//w:tag")
                      title_elements = prop.xpath(".//w:alias")
  
                      ns_val = (
                          "{http://schemas.openxmlformats.org/wordprocessingml/"
                          "2006/main}val"
                      )
                      tag = tag_elements[0].get(ns_val, "")
                      title = title_elements[0].get(ns_val, "")
  
                      controls.append({
                          "tag": tag,
                          "title": title,
                      })
  
          return {
              "status": "success",
              "filepath": filepath,
              "content_controls": controls,
              "count": len(controls),
          }

• src/docx_mcp/server.py:435-435 (registration)

  The tool is registered using the @app.tool() decorator.

  @app.tool()