you can already do that for most things using the Wingman Core API and/or websocket but this is a "dev thing". I guess you mean doing that in the UI? Question is what exactly would you be able to do and how complicated would that be for users to setup? We don't want to bloat Wingman AI if we can avoid it.