You seem to have both scripts running on connect. The on_disconnect (kagu_stop) should have RemoteDeviceDisconnect, not RemoteDeviceConnect.

As far as I can tell this is not a dbus-scripts-setting bug. Are you sure you selected disconnect for the second line?

About getting the same dbus signal for connect/disconnect - there are other such events, such as keyboard slide (same signal when openned and closed) USB device connect/disconnect. In such cases you need to have the script running on both events, and the script should use some other method for finding out which event actually happened (and be really careful of race conditions).