Unfortunately this can’t be fixed via robots.txt — Google won’t proactively crawl the URL, but it does still index pages it finds from external links.
A `noindex` meta tag, however, does prevent pages from being indexed, so that feels like the appropriate solution here.
Somehow this keeps happening, followed by a rediscovery of this ancient old knowledge.
And you're talking about relieving a symptom on a well-behaved and well-documented public service. Getting indexed by Google indicates that a link has already been crawled, indexed, and scraped by a lot of services, most of which cannot be seen or controlled by ordinary mortals.