So far it looks like Twitter is maintaining some type of database of URLs of news stories, etc. that are known disinfo and auto-flagging any tweets that contain the url. There appear to manual processes for highly visible users like Trump.
Also, people can report, so they are