Try Locally

Run the agent against a throwaway KinD cluster to see what NudgeBee looks like with real data, without touching a real cluster.

What you get, and what you don't

KinD nodes are containers sharing the host kernel, so the eBPF node agent generally cannot attach its probes. The install below turns it off. You still get workload inventory, events, metrics, and recommendations. You do not get network metrics, L7 traces, or profiling — those need a real node.

Prerequisites

• Docker
• kubectl
• KinD
• Helm

1. Create the cluster

kind create cluster --config <(cat <<EOF
kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
nodes:
- role: control-plane
- role: worker
- role: worker
EOF
)

kubectl cluster-info
kubectl get nodes

2. Get an auth key

In app.nudgebee.com, go to Admin → Integrations, open the Kubernetes Clusters card, and click Add K8s Account. Name it something you will recognise as disposable, mark it Non-production, and finish the wizard. Copy the auth key it gives you.

3. Install the agent

wget https://raw.githubusercontent.com/nudgebee/k8s-agent/main/installation.sh
sh installation.sh -a <agent-key> -d true

-d true disables the node agent, which is what you want on KinD. The script installs everything else, including a Prometheus stack if the cluster has none.

4. Wait

Inventory and metrics show up in the UI within about 5 minutes. Rightsizing and cost recommendations need roughly an hour of history before they appear — a fresh cluster has nothing to recommend from yet.

kubectl get pods -n nudgebee-agent

5. Clean up

kind delete cluster

Also delete the Kubernetes account in Admin → Integrations → Kubernetes Clusters, otherwise it lingers in the UI as a cluster that stopped reporting.