Google Workspace MCP Server

Instructions

Implementation Reference

• gdrive/drive_tools.py:91-175 (handler)

  The main asynchronous handler function that downloads and extracts text content from a Google Drive file by ID. Supports Google native formats via export, Office XML parsing, and UTF-8 decoding with binary fallback. Includes file metadata header in response.

  async def get_drive_file_content(
      service,
      user_google_email: str,
      file_id: str,
  ) -> str:
      """
      Retrieves the content of a specific Google Drive file by ID, supporting files in shared drives.
  
      • Native Google Docs, Sheets, Slides → exported as text / CSV.
      • Office files (.docx, .xlsx, .pptx) → unzipped & parsed with std-lib to
        extract readable text.
      • Any other file → downloaded; tries UTF-8 decode, else notes binary.
  
      Args:
          user_google_email: The user’s Google email address.
          file_id: Drive file ID.
  
      Returns:
          str: The file content as plain text with metadata header.
      """
      logger.info(f"[get_drive_file_content] Invoked. File ID: '{file_id}'")
  
      file_metadata = await asyncio.to_thread(
          service.files().get(
              fileId=file_id, fields="id, name, mimeType, webViewLink", supportsAllDrives=True
          ).execute
      )
      mime_type = file_metadata.get("mimeType", "")
      file_name = file_metadata.get("name", "Unknown File")
      export_mime_type = {
          "application/vnd.google-apps.document": "text/plain",
          "application/vnd.google-apps.spreadsheet": "text/csv",
          "application/vnd.google-apps.presentation": "text/plain",
      }.get(mime_type)
  
      request_obj = (
          service.files().export_media(fileId=file_id, mimeType=export_mime_type)
          if export_mime_type
          else service.files().get_media(fileId=file_id)
      )
      fh = io.BytesIO()
      downloader = MediaIoBaseDownload(fh, request_obj)
      loop = asyncio.get_event_loop()
      done = False
      while not done:
          status, done = await loop.run_in_executor(None, downloader.next_chunk)
  
      file_content_bytes = fh.getvalue()
  
      # Attempt Office XML extraction only for actual Office XML files
      office_mime_types = {
          "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
          "application/vnd.openxmlformats-officedocument.presentationml.presentation",
          "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
      }
  
      if mime_type in office_mime_types:
          office_text = extract_office_xml_text(file_content_bytes, mime_type)
          if office_text:
              body_text = office_text
          else:
              # Fallback: try UTF-8; otherwise flag binary
              try:
                  body_text = file_content_bytes.decode("utf-8")
              except UnicodeDecodeError:
                  body_text = (
                      f"[Binary or unsupported text encoding for mimeType '{mime_type}' - "
                      f"{len(file_content_bytes)} bytes]"
                  )
      else:
          # For non-Office files (including Google native files), try UTF-8 decode directly
          try:
              body_text = file_content_bytes.decode("utf-8")
          except UnicodeDecodeError:
              body_text = (
                  f"[Binary or unsupported text encoding for mimeType '{mime_type}' - "
                  f"{len(file_content_bytes)} bytes]"
              )
  
      # Assemble response
      header = (
          f'File: "{file_name}" (ID: {file_id}, Type: {mime_type})\n'
          f'Link: {file_metadata.get("webViewLink", "#")}\n\n--- CONTENT ---\n'
      )
      return header + body_text

• gdrive/drive_tools.py:88-90 (registration)

  Registers the tool with the MCP server using @server.tool(), applies HTTP error handling decorator with tool name, and requires Google Drive read authentication.

  @server.tool()
  @handle_http_errors("get_drive_file_content", is_read_only=True, service_type="drive")
  @require_google_service("drive", "drive_read")

• gdrive/drive_tools.py:91-109 (schema)

  Function signature with type annotations and comprehensive docstring defining input parameters (user_google_email: str, file_id: str) and return type (str), describing tool behavior and supported file types.

  async def get_drive_file_content(
      service,
      user_google_email: str,
      file_id: str,
  ) -> str:
      """
      Retrieves the content of a specific Google Drive file by ID, supporting files in shared drives.
  
      • Native Google Docs, Sheets, Slides → exported as text / CSV.
      • Office files (.docx, .xlsx, .pptx) → unzipped & parsed with std-lib to
        extract readable text.
      • Any other file → downloaded; tries UTF-8 decode, else notes binary.
  
      Args:
          user_google_email: The user’s Google email address.
          file_id: Drive file ID.
  
      Returns:
          str: The file content as plain text with metadata header.