To answer this question need to some environment related details, like are you using cloud orchestrator and AI center, or the setup is in on-premise servers.
also if you want to try quickly you can create an public endpoint for you ML skill and you can use that endpoint directly in the ML related activities like ML extractor.