Once you have a lot of services and a lot of people touching things over time, it gets surprisingly hard to tell which alarms still make sense.
I ended up hacking together a small, read-only CLI that just lists alarms across regions and highlights those kinds of cases.
This isn’t meant to fix anything automatically — mostly a way to see what you actually have. Curious if this matches what others see in real setups, or if you approach this problem differently.