Query traffic and compute for one app, a whole workspace, or everything you run: requests, unique users, status codes, CPU time, memory, and network use, in buckets from five minutes to a week.
A dashboard for one app. Its requests, unique users, and CPU time add up through the day, and each new request shows in the totals, the chart, and the log.
An illustration, with example data.
Powering products and agents at
Pass an app, a start, and an end. Read the totals for the whole window, and the grouped results for the trend.
$ npm install stackmachineimport StackMachine from "stackmachine";
const client = new StackMachine(process.env.STACKMACHINE_API_KEY);
const usage = await client.usage.metrics({
app: "da_XYZ",
start: "2026-10-01T00:00:00Z",
end: "2026-10-08T00:00:00Z",
groupedBy: "BY_DAY",
});
const { requests, workloads } = usage.totals;
console.log(requests.totalRequests, requests.uniqueUsers);
console.log(workloads.realCpuTimeMillis, workloads.networkEgressBytes);
for (const day of usage.grouped) {
console.log(day.groupedAt, day.requests.totalRequests);
}Metrics show what changed. Logs show why.
Logs are kept per app version. Fetch them by time range, filter by stream (STDOUT, STDERR, or RUNTIME), search the text, or narrow them to one instance.
Pass owner instead of app to add up everything a user or workspace runs. Leave both out to cover every resource you can see.
| app | requests | CPU time |
|---|---|---|
| storefront | 12,480 | 38.2 s |
| admin | 1,926 | 9.4 s |
| webhooks | 804 | 2.1 s |
More in the documentation.
One API key for hosting, storage, databases, email, and cron.