k6 script with realistic browse-vs-buy traffic and thresholds wired as a CI gate, ready to run

Load-testing suite for a checkout API using k6. Models a realistic mixed-traffic profile (browsers, buyers, bots) with proper ramp and pass/fail thresholds wired as CI gates. ```javascript import http from "k6/http"; import { check, sleep } from "k6"; import { Trend } from "k6/metrics"; const checkoutLatency = new Trend("checkout_latency", true); export const options = { scenarios: { browse: { // 80% just browsing executor: "ramping-vus", startVUs: 0, stages: [ { duration: "2m", target: 200 }, // ramp up { duration: "5m", target: 200 }, // sustained peak { duration: "1m", target: 0 }, // ramp down ], exec: "browse", }, checkout: { // 20% convert executor: "constant-arrival-rate", rate: 30, timeUnit: "1s", duration: "8m", preAllocatedVUs: 50, exec: "checkout", }, }, thresholds: { http_req_failed: ["rate<0.01"], // <1% errors or the build fails checkout_latency: ["p(95)<500", "p(99)<1000"], }, }; export function browse() { http.get("https://staging.shopflow.io/api/products"); sleep(Math.random() * 3); // think time } export function checkout() { const res = http.post("https://staging.shopflow.io/api/orders", JSON.stringify({ sku: "SKU-004521", qty: 1 }), { headers: { "Content-Type": "application/json" } }); checkoutLatency.add(res.timings.duration); check(res, { "order created": (r) => r.status === 201 }); sleep(1); } ``` **Scenarios covered:** baseline (200 VUs), peak (the sustained stage), and a spike variant (swap in a stage jumping to 1000 VUs over 30s). For a **soak test** rerun with the sustained stage at `8h`. **CI gate:** `k6 run --quiet load.js` — the thresholds make k6 exit non-zero on a regression, so a slow build blocks the deploy automatically.