To help with preventing the tool from creating empty or almost empty archieves, the tool could write out number of resources of each type (could be just CRDs or all types of resources) and within each namespace that were captured. This immediately brings users attention to the right namespaces (if the counts of resources are surprising) and is transparent when it comes to what is being captured.