Ideally we would measure Request minus Max_Usage_Over_7_Days. Since this is a snapshot tool (wrapping kubectl top) it can't see historical peaks so it def lean

Ideally we would measure Request minus Max_Usage_Over_7_Days. Since this is a snapshot tool (wrapping kubectl top) it can't see historical peaks so it def leans towards being a current state audit. Love the idea of pairing this with a load test benchmark that would actually give you a calculated Safe Request Size rather than just a guess.