What problem does it solve? Systems fail under real traffic when capacity is never validated, bottlenecks remain hidden, and performance regressions slip into production. This Skill provides structured workflows for load testing, profiling, and optimization so you can find breaking points, diagnose slow code paths, and enforce performance SLOs before users are affected. ## Core Features & Use Cases - Load, Stress, Soak, and Spike Testing: Ready-to-run k6 (JavaScript) and Locust (Python) test templates with thresholds that fail CI builds when latency or error budgets are violated. - Cross-Language Profiling: CPU, memory, and I/O profiling workflows using py-spy and memory_profiler for Python, pprof for Go, and Chrome DevTools or clinic.js for Node.js. - Optimization Playbooks: Concrete patterns for caching with Redis, eliminating N+1 database queries, cursor-based API pagination, and improving Core Web Vitals (LCP, INP, CLS). - Use Case: Before a product launch, run a k6 stress test to find the API's breaking point, profile the hottest endpoints with py-spy, add caching and query indexes, then wire the test into GitHub Actions to catch future regressions. ## Quick Start Ask the agent to write and run a k6 load test against your API endpoint with a p95 latency threshold under 500ms.
