Reddit

In the Libreddit Matrix chat I've seen them talk about the GraphQL API as well as webscraping, but there are problems/potential problems with both. But I think because libreddit doesn't do anything with user accounts there is at least a better chance of it surviving, at least after the initial changes.