We currently have alerts for APIRemovedInNextReleaseInUse and APIRemovedInNextEUSReleaseInUse to tell folks when they may need to update a client before they update their cluster to the next (EUS) release. However, it seems theoretically possible for an API to be deprecated without committing to a specific future removed-in version, or for a committed removed-in version to be out beyond the next EUS release. If that sort of thing happens in practice, it may be worth creating a new DeprecatedAPIInUse alert or otherwise poking the workload admins so they have plenty of lead time to work on updating their clients. In the absence of a way to poke workload admins, a cluster-scoped alert would poke cluster admins (and CI suites, etc.), and then the cluster admins (and CI monitors, etc.) could track down and inform relevant workload admins. But the end goal is getting the word out to workload admins as conveniently and early as possible, so client-updating can proceed calmly and slowly, without any "oh no, next release this is gone?!" drama.