Dashboard

The Supernode dashboard is the intended operator entrypoint for workload lifecycle and monitoring.

The preferred overall management experience is still agent-first: load the Supernode skills into your agent, then use the dashboard as the main visual and operational surface.

It is meant to make the Supernode operating model visible:

discover supported workloads
install them into isolated namespaces
guide onboarding flows
inspect workload health and logs
jump into Grafana for deeper analysis

What the dashboard depends on

The dashboard is part of the control-plane and expects the rest of the platform to exist.

At minimum, the dashboard environment depends on:

PROMETHEUS_ENDPOINT
REGISTRY_ENDPOINT
OCI_ENDPOINT
GRAFANA_API_ENDPOINT
GRAFANA_PUBLIC_URL

These are the canonical integrations that let the dashboard query workload health, search OCI-packaged extensions, install them, and resolve Grafana dashboards.

Operator workflow

The intended operator experience is:

open the dashboard after bootstrap
review currently installed workloads
add a new workload from the available catalog
complete any workload-specific onboarding flow
inspect health, logs, and linked dashboards

For local dashboard access through an agent-guided workflow, use supernode-dashboard-port-forward.md.

Catalog and installation model

The dashboard discovers supported workloads from the configured OCI registry.

When an operator adds a workload, the platform is expected to:

create a namespace for that workload
install the corresponding Helm chart
apply the Supernode status model to the workload
expose it in the dashboard overview

Workload detail view

Each workload detail page is intended to give operators a single place to inspect:

current status
historical health
live logs
workload-specific telemetry
linked Grafana dashboards

For Cardano-style node workloads, the detail page is also expected to surface chain, peer, propagation, and producer metrics.