It seems like there's a generic user agent python-requests/2.31.0 which is overusing the copyvios API. In the past day, they account for 502 out of the 3,288 requests with the search engine enabled. Another UA, Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36 accounts for 298. Both were using the API for requests. They were running checks on French and Hebrew Wikipedia, respectively.

There should probably be a rate limit or login or API requests, and specifically a ban for generic user agents trying to use the search engine. We're hitting 429s again even with the login requirements that were placed a few months back.