Introduction
Introduction
What's Kubeadapt?
Kubeadapt is a read-only Kubernetes cost monitoring and optimization platform. A lightweight agent reads your cluster's resource state and usage, ships it to Kubeadapt Cloud for analysis, and surfaces cost attribution down to the container, rightsizing recommendations for every workload, and idle-resource detection.
The agent never modifies your cluster. You see exactly what to fix; you decide when and how to apply changes.
Supported providers
Cloud-managed, self-hosted, or somewhere in between — the agent runs anywhere Kubernetes does.
Cloud-managed Kubernetes
- Amazon EKSFull support
Including Reserved Instances and Savings Plans via the AWS account integration.
- Google GKEOn-demand + spot
On-demand and spot pricing today; Committed Use Discount integration is planned.
- Azure AKSOn-demand + spot
On-demand and spot pricing today; Reserved Instance and Savings Plan integration is planned.
Self-hosted
kubeadm,kops, RancherFull support- Red Hat OpenShiftFull support
- Bare-metal and on-premisesCustom pricing
Override the default per-core and per-GiB rates from the dashboard.
Every cluster needs Kubernetes 1.24+, the standard metrics-server, and outbound HTTPS to ingest.kubeadapt.io.
How to Get the Most Out of Kubeadapt
Three modes, in the order most teams adopt them.
Detect
Connect your first cluster, wait roughly 24 hours for usage data to accumulate, then open the Dashboard. Look at total spend, abandoned workloads, and idle nodes.
Surfaces 10–20% of cost waste.
Optimize
Open Available Savings. Apply CPU and memory rightsizing recommendations to non-critical workloads first; promote spot-eligible workloads to spot.
Week-one savings of 20–40% are typical.
Prevent
Turn on Best Practices checks. Enable the optional eBPF agent for cross-AZ network cost visibility.
Shift-Left cost gates coming next.
How to get started in 5 minutes or less
- 1
Add the Helm repository.
helm repo add kubeadapt … - 2
Generate an agent token.
Open app.kubeadapt.io → Clusters → Add Cluster. Choose your provider and pick production-like or non-production-like.
- 3
Install with Helm.
Use the generated command for your cluster.
- 4
Verify the agent.
Check the pod is healthy:
kubectl get pods -n kubeadapt
1# 1. Add the Helm repository
2helm repo add kubeadapt https://kubeadapt.github.io/kubeadapt-helm
3helm repo update
4
5# 2. Generate an agent token at app.kubeadapt.io → Clusters → Add Cluster
6# Pick your provider, choose production-like or non-production-like.
7
8# 3. Install
9helm install kubeadapt kubeadapt/kubeadapt \
10 --namespace kubeadapt \
11 --create-namespace \
12 --set agent.config.token="<your-token>"
13
14# 4. Verify the agent pod is healthy
15kubectl get pods -n kubeadaptCost data starts flowing within minutes; full breakdowns appear in one to twenty-four hours depending on cluster activity.
Where to get help?
Five ways to reach us — pick whichever fits the question.
Email support
Real engineers respond, usually within a business day.
FAQ
Read-only model, supported clusters, data handling, billing.
API & CLI
Programmatic access via the REST API or the kubeadapt CLI.
Status
System uptime, incidents, scheduled maintenance.
Feedback
Vote on features, request integrations, file ideas.