Oof, this is exactly the nightmare scenario for “repo-first” OSS.

The weird bit isn’t that a scraper site exists, it’s that Google can’t do the obvious graph join: query == project name, #1 result is the repo, repo declares Homepage = X, yet Google still boosts an imposter domain. That’s not “SEO”, that’s the ranking system refusing to treat maintainer-declared canonical as a strong signal. Early domain squatters get to “set the default” purely by being first, then they can flip the content later once trust is baked in.

People keep saying “tell users to bookmark the real URL” like that scales. Most people will click the second link and assume it’s official. If Google can’t solve this class of problem, their “AI answers” are going to be a bigger mess than blue links ever were.