load-profile-generator·codegen·active

Load Profile Generator

Translates business-level load expectations (peak Black Friday traffic, end-of-month reporting) into k6/JMeter/Locust scripts with realistic think times.

LEOtrigger · agent

Translates business-level load expectations into k6, JMeter, or Locust scripts with realistic think times, ramp curves, and traffic mix.

When this skill fires

·A user describes load in natural language:"simulate Black Friday peak for our checkout flow"

·A capacity-planning ticket references a target throughput and SLO

·An ARIA recommendation calls for load testing as part of a release readiness check

Translation layers

Business → traffic shape

·"Black Friday peak" → 12× baseline RPS over 4 hours, with a 30-minute spike at hour 1

·"End-of-month reporting" → sustained 3× baseline for 6 hours

·"Marketing campaign launch" → 2× ramp over 30 min, hold for 3 hours, ramp down

LEO maintains a small library of these "shapes" derived from real customer load profiles.

Traffic shape → user journey mix

The shape alone isn't enough — Black Friday checkout traffic looks different from Black Friday browsing traffic. LEO infers a journey mix from the requirement:

·Browse (50%)

·Add to cart (25%)

·Checkout (15%)

·Account / login (10%)

Mix can be overridden when the user provides one explicitly.

Journey mix → script

LEO emits a script in the chosen framework with:

·A virtual user definition per journey

·Realistic think times between actions (sampled from a distribution, not constant)

·Authentication setup and token reuse where appropriate

·Tagged metrics soThreshold Detector can analyze per-journey

Realistic think times

→Constant think times (sleep(1) between every action) are unrealistic and produce optimistic results. LEO uses log-normal distributions calibrated to the journey type — a checkout flow has longer pauses than a browse flow.

Ramp curves

·Linear — for capacity planning, default

·Stepped — for finding breaking points, increases users in discrete jumps

·Spike — sudden 10×+ load to test resilience

·Soak — flat load for hours to surface memory leaks and resource exhaustion

What it produces

A complete script file (k6 by default):

·Imports and setup

·VU groups per journey, with thinkable times and assertions

·Stages array defining the ramp curve

·Custom metrics for journey-level latency and error rate

·A README documenting the assumed traffic mix and how to override

→The README is mandatory. A load test you can't reason about in two minutes is worse than no load test.

Related · codegen

Step Definition Generator

step-def-generator

Converts Gherkin steps into Cucumber/TypeScript step definitions. Wires parameters, expression types, and shared world state automatically.

Page Object Generator

page-object-generator

Inspects DOM (live or via spec) and emits typed Page Objects with stable selectors and high-level actions. Enforces single-responsibility per object.

Locator Strategist

locator-strategist

Picks the most stable selector for a target element — role-based first, label-based next, test-id last. Avoids brittle CSS class chains.

API Test Generator

api-test-generator

Reads OpenAPI / GraphQL schemas and emits contract + integration tests. Handles auth, pagination, idempotency, and rate-limit assertions.

Test Data Factory

test-data-factory

Synthesizes realistic test data — names, addresses, payment cards, medical records, account numbers — with deterministic seeds and PII safety.

Assertion Expander

assertion-expander

Augments thin Gherkin Then-clauses with additional assertions: response codes, side-effects, audit log entries, downstream events.