Clusters tend to accumulate garbage: debug pods, failed jobs, test workloads that nobody cleans up. I built Columbo, a CLI that tries to detect these by assigning a suspicion score to each pod.
Would love feedback.