Presently there's only the GUI to run the schema parser. This is very convenient - especially because it remembers output directories and stuff.
It was originally planned that the parser could be run as a command line application with switches to set what outputs, directories and stuff.
This is still to be done.