- COST
- PODS
- TRENDS
- HUMAN
Short answer
How can I find out which Kubernetes namespaces are costing me the most?
Your AI uses the OpenCost connection to pull cost allocations for every namespace and pod. It identifies the biggest spenders and provides a granular breakdown of your cluster usage.
Where your data lands.
The outcomes of your audit.
The AI processes raw cluster data into these specific results.
COST
Namespace spend totals
The AI pulls the exact dollar amount spent per namespace. You get a clear list of your most expensive environments.
PODS
Granular pod breakdown
Your agent drills down from the namespace level to specific pods. This shows you exactly which workloads are driving up your bill.
TRENDS
Usage patterns
The AI looks for spikes in resource consumption. You'll see which namespaces are growing in cost over time.
HUMAN
Optimization decisions
The AI presents the data, but you decide which pods to resize or terminate. You keep full control over your cluster configuration.
The workflow
What your AI does when the data arrives.
The AI acts as a data analyst for your Kubernetes cluster.
Fetch cluster data
The AI pulls the current cost distribution across your entire cluster.
get_allocationGroup by namespace
It aggregates the raw data to show you the total cost for every namespace.
get_allocationIdentify outliers
The agent flags namespaces or pods that are consuming disproportionate amounts of resources.
get_allocationSummarize findings
The AI translates the technical cost metrics into a plain English report for you.
get_allocation
Try it
Copy these to start.
Use these prompts to begin your audit.
Starting points
These are opening lines. Swap out the namespace names or cluster IDs to match your environment.
OpenCost Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_5JgTTx4dCCg1vy5LoQY7EOXCEeiHxNr943mMQtmk/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — OpenCost capabilities are ready to use.
{
"mcpServers": {
"opencost-k8s-cost-mcp": {
"url": "https://edge.vinkius.com/vk_preview_5JgTTx4dCCg1vy5LoQY7EOXCEeiHxNr943mMQtmk/mcp"
}
}
}Show me a breakdown of costs by namespace in my current cluster.
Which pods in the 'production' namespace are driving the highest costs?
List the top 5 most expensive namespaces right now.
Compare the cost of the 'staging' namespace against the 'development' namespace.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect OpenCost once, then ask.
Just link your OpenCost instance to Vinkius. Your credentials stay encrypted, and you can immediately start asking your AI about your cluster spend.
Connect OpenCost to your AIFAQ
How this task behaves.
- 01
Can the AI change my Kubernetes resource limits?
No. The AI can only read cost data through the OpenCost connection. You must manually adjust limits in your cluster.
- 02
Does the AI see my actual application code?
No. The AI only accesses the cost and resource allocation data provided by OpenCost.
- 03
Can I ask for cost trends over the last month?
Yes, as long as your OpenCost instance has that historical data available for the AI to read.
- 04
Will the AI delete any pods to save money?
No. The AI has no permission to modify, delete, or create any resources in your cluster.
- 05
How often does the AI refresh the cost data?
The AI pulls fresh data from OpenCost every time you ask a question.
More questions about OpenCost? The Connector page answers them. See everything the OpenCost Connector can do
Connect OpenCost to Claude, Cursor, ChatGPT & more
