CSV MCP Server

Instructions

Remove a specific row from the CSV file.

Args:
    filename: Name of the CSV file
    row_index: Zero-based index of the row to remove

Returns:
    Dictionary with removal results

Implementation Reference

• csv_mcp_server/server.py:128-144 (handler)

  MCP tool handler for 'remove_row', registered via @mcp.tool() decorator. Delegates execution to CSVManager instance and handles exceptions.

  @mcp.tool()
  def remove_row(filename: str, row_index: int) -> Dict[str, Any]:
      """
      Remove a specific row from the CSV file.
      
      Args:
          filename: Name of the CSV file
          row_index: Zero-based index of the row to remove
      
      Returns:
          Dictionary with removal results
      """
      try:
          return csv_manager.remove_row(filename, row_index)
      except Exception as e:
          return {"success": False, "error": str(e)}

• csv_mcp_server/csv_manager.py:260-290 (helper)

  Core implementation of row removal logic in CSVManager class: validates file, creates backup, loads DataFrame with pandas, drops the specified row, saves back to CSV, returns success info with removed row data.

  def remove_row(self, filename: str, row_index: int) -> Dict[str, Any]:
      """Remove a specific row from the CSV file."""
      filepath = self._get_file_path(filename)
      
      if not filepath.exists():
          raise FileNotFoundError(f"CSV file '{filename}' not found")
      
      # Create backup
      self._create_backup(filepath)
      
      try:
          df = pd.read_csv(filepath)
          
          if row_index >= len(df):
              raise IndexError(f"Row index {row_index} out of range (max: {len(df)-1})")
          
          removed_row = df.iloc[row_index].to_dict()
          df = df.drop(df.index[row_index])
          
          df.to_csv(filepath, index=False)
          
          logger.info(f"Removed row from CSV file: {filepath}")
          return {
              "success": True,
              "filename": filename,
              "removed_row": removed_row,
              "new_total_rows": len(df)
          }
      except Exception as e:
          logger.error(f"Failed to remove row: {e}")
          raise