performance-test

Design load, stress, and endurance tests with k6, JMeter, and Locust.

1|Updated Mar 21, 2026
One-click install
npx skills add https://github.com/kalilurrahman/kr-claudiator-skills-original-prompts --skill performance-test-kalilurrahman
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: performance-test
Source: https://github.com/kalilurrahman/kr-claudiator-skills-original-prompts/tree/main/07-testing-quality/performance-test
Command: npx skills add https://github.com/kalilurrahman/kr-claudiator-skills-original-prompts --skill performance-test-kalilurrahman

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often ship applications without knowing how they behave under real traffic, leading to outages, SLA violations, and undetected bottlenecks. This Skill provides a systematic process for designing performance tests with defined SLAs, realistic scenarios, and bottleneck analysis instead of ad-hoc testing. ## Core Features & Use Cases - Test Scenario Design: Covers load, stress, spike, and endurance test types with ramp-up patterns, think time, and realistic user data. - Multi-Tool Support: Provides ready-to-adapt examples for k6 (JavaScript), JMeter (GUI/CLI/distributed), and Locust (Python) with thresholds and custom metrics. - Bottleneck Diagnosis & Capacity Planning: Maps symptoms like high response time or memory growth to root causes, plus autoscaling and capacity calculations. - Use Case: Before a product launch expecting 1000 concurrent users, generate a k6 script with p95 < 200ms thresholds, run it in CI, and produce a report identifying the database connection pool as the bottleneck. ## Quick Start Ask the AI to design a k6 load test for your API with a p95 under 200ms SLA at 1000 concurrent users, including bottleneck analysis and a capacity plan.

Frequently Asked Questions about performance-test

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a load test with k6?

Write a JavaScript file exporting an options object with stages for ramp-up and thresholds like p(95)<200ms, then define a default function making HTTP requests with checks. Run it with the k6 run command and review the summary metrics output.

k6 vs JMeter vs Locust: which load testing tool should I use?

k6 suits modern teams wanting JavaScript scripting and CI integration, JMeter fits enterprise environments needing a GUI and distributed testing, and Locust works well for Python teams defining user behavior as code with task weights.

How do I integrate performance tests into CI/CD?

Use the grafana/k6-action GitHub Action to run your k6 script on pull requests, then add a step that parses the results JSON and fails the build if p95 response time exceeds your SLA threshold.

Why does my error rate spike at a specific number of users?

A sudden error spike at an exact user count usually indicates connection pool exhaustion. Check database metrics like Threads_connected versus max_connections, then increase the pool size or add queueing.

How long should a performance test run to catch memory leaks?

Run endurance tests for extended periods, ideally hours rather than minutes, since short 5-minute tests miss memory leaks and resource exhaustion. Monitor memory growth over time and profile with heap dumps if it climbs steadily.