Skip to main content

Usage

1. Overview

The Usage screen is the Nukk platform's FinOps panel: a detailed breakdown of every cost generated by your account, resource by resource, with granularity down to CPU, memory, network, execution time, and even AI token cost. Instead of finding out the cost only on the invoice at the end of the month, you can track in (near) real time what's consuming credit, where, and why — and act before the cost gets out of control.

2. Filters and Indicators

At the top, two filters narrow down the view:

FilterDescription
MonthReference month for consumption (e.g. 2026-07).
EnvironmentRestricts the view to a specific environment, or All environments to see them all combined.

Right below, four indicators summarize the selected period:

IndicatorDescription
Total this monthTotal accumulated cost for the month, in credits (cr).
ResourcesNumber of resources that generated cost in the period.
Tokens costSpecific cost of AI token consumption (e.g. Copilot, vibe code), in credits.
vs last monthPercentage change in total cost compared to the previous month (e.g. -72.3%).

The Refresh button updates the data shown, and the Last update time shows when the numbers were last calculated.

3. Cost Trend

A trend chart shows the evolution of total cost over the last 6 months, making it easy to spot consumption spikes, gradual growth, or drops — for example, a peak between 2026-05 and 2026-06 followed by a drop in 2026-07.

4. Resources — Breakdown by Resource

Below the chart, the Resources table breaks down cost resource by resource, with filter tabs by type:

TabWhat it shows
AllAll resources that generated cost.
FlowsCost generated by Flows.
FrontendCost generated by Frontend Apps.
BackendCost generated by Backend Apps.
AddonsCost generated by deployed Add-ons.
PreviewsCost generated by application preview environments.
TokensCost generated by AI usage (tokens consumed).

Each row in the table shows:

ColumnDescription
NameResource name (e.g. nukk-site).
TypeResource type (e.g. frontendapp, addon, preview).
SizeResource size/plan, when applicable.
EnvironmentEnvironment the resource ran in (e.g. Test).
CPU (core·h)Accumulated CPU consumption, in core-hours.
Memory (GiB·h)Accumulated memory consumption, in GiB-hours.
Network (GiB)Network traffic consumed, in GiB.
Running (h)Total time the resource was running, in hours.
CreditsTotal cost generated by that resource in the period, in credits.

This lets you answer questions like "which resource is weighing the most on the bill?" (e.g. comparing nukk-site, its nukk-site-tunnel tunnel add-on, and the nukk-site preview side by side) without having to manually cross-reference information.

5. Why This Matters

  • For developers: shows the real impact of each Flow, App, or Add-on on cost — useful for spotting oversized resources (e.g. memory allocated far above what's needed) before it becomes a recurring cost problem.
  • For project/squad managers: provides FinOps visibility without relying on spreadsheets or the cloud provider's invoice — month-over-month comparison, by environment, by resource type, all within the platform itself.
  • For architecture decisions: separating infrastructure cost (CPU/memory/network) from AI token cost helps understand whether spend comes from "running the application" or "using AI to build/operate the application" — two types of cost with very different optimizations.

6. FAQ

What counts as "Resources" in the top indicator? TODO: confirm the exact definition (distinct resources in the period vs. executions).

Can Usage data be exported (e.g. CSV) for external reports? TODO: confirm whether there's export beyond the on-screen visualization.

Is Tokens cost billed separately from infrastructure cost? TODO: confirm whether they appear together on the same invoice or on separate line items.