I don't think there's a good reason. I've opened a PR to fix it.

I suspect this was done originally because the flag was intended for internal use, when KRR is run by Robusta runner in-cluster. (As opposed to the usual flow where KRR is run outside of the cluster.)

But there is no reason not to expose it more broadly if someone wants to run KRR standalone inside an openshift cluster.