Cluster dashboard

A cluster-wide overview in a single screen: a health banner that grades the cluster, node and workload status, live CPU and memory, and your heaviest pods - kept current as the cluster changes, with no separate metrics backend to stand up. Show, hide and reorder the sections; your layout is saved per cluster.

• A health banner that grades the cluster - Healthy, Needs attention or Critical
• Node and workload status: Ready/NotReady, cordoned and pressured nodes, degraded deployments
• Live aggregate CPU and memory, with capacity headroom and requested-vs-used coverage
• Your heaviest pods by CPU and memory, surfaced first - each with its namespace
• Active problems pulled to the top - the exact nodes and workloads that need you
• Show, hide and reorder sections; the layout is saved per cluster