performance-engineer

Diagnose and optimize application performance using observability, profiling, and load testing.

1|Updated May 10, 2026
One-click install
npx skills add https://github.com/Tgoldi/claude-skills --skill performance-engineer-tgoldi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: performance-engineer
Source: https://github.com/Tgoldi/claude-skills/tree/main/performance-engineer
Command: npx skills add https://github.com/Tgoldi/claude-skills --skill performance-engineer-tgoldi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Slow applications, hidden bottlenecks, and unmonitored systems degrade user experience and inflate infrastructure costs. This Skill provides a structured performance engineering workflow to measure baselines, isolate bottlenecks, and implement validated optimizations across frontend, backend, and infrastructure layers. ## Core Features & Use Cases - Observability & Monitoring: Set up OpenTelemetry distributed tracing, Prometheus/Grafana metrics, APM platforms, and Real User Monitoring with Core Web Vitals tracking. - Profiling & Load Testing: Analyze CPU, memory, and I/O with flame graphs and heap analysis, then validate with k6, JMeter, Gatling, or Locust load tests. - Multi-Tier Optimization: Optimize caching layers (Redis, CDN, browser), database queries, frontend bundles, and cloud auto-scaling configurations. - Use Case: Your API latency spiked after a release. Use this Skill to correlate distributed traces, profile the slow endpoints, propose caching and query optimizations, and set performance budgets to prevent regression. ## Quick Start Ask the assistant to analyze your application's performance bottlenecks and propose an optimization plan with monitoring and load testing.

Frequently Asked Questions about performance-engineer

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

FAQPage Schema
How do I diagnose performance bottlenecks in a web application?

Start by establishing a baseline with metrics and profiling, then use distributed tracing to isolate slow services or endpoints. This Skill guides CPU, memory, and I/O profiling with flame graphs and heap analysis to pinpoint the highest-impact bottlenecks first.

What load testing tools should I use for API performance testing?

k6, JMeter, Gatling, Locust, and Artillery are all supported for load testing REST, GraphQL, and WebSocket APIs. The Skill helps design realistic traffic scenarios, set performance budgets, and integrate regression detection into CI/CD pipelines.

How do I set up OpenTelemetry distributed tracing?

OpenTelemetry instrumentation collects traces and metrics across services, which can be exported to platforms like Jaeger, Honeycomb, Prometheus, or Grafana. The Skill covers correlation across services, SLI/SLO tracking, and alerting configuration.

Can this help improve Core Web Vitals scores?

Yes, it covers LCP, FID, and CLS optimization through image optimization, lazy loading, bundle splitting, critical CSS, and HTTP/2 or HTTP/3 delivery. It also sets up Real User Monitoring to track user-perceived performance over time.

When should I not run load tests against production?

Avoid load testing production without explicit approvals and safeguards such as traffic isolation and rate limits. The Skill recommends staging environments, staged rollouts, and rollback plans for high-risk performance changes.

What caching strategy works for high-traffic applications?

Multi-tier caching combines application-level in-memory caches, distributed stores like Redis or Memcached, database query caching, CDN edge caching, and browser cache headers. The Skill helps choose layers and define invalidation strategies per workload.