As of right now, we don’t support parsing images, since the bot wouldn’t know what to do with that information coming in, as all incoming information is parsed as a UserMessage and expects test for parsing, not a url or bytes.
You could extend the facebook channel, so that here:
if there is an image, you can do something else, e.g. send an /out_of_scope message to the bot so that it can respond that the user’s request is out of scope