performance-engineering — Claude Code skill: setup and usage
performance-engineering — a skill for Claude Code designed for full-stack web application performance optimization. Activates when Core Web Vitals drop, bundle sizes increase, or page response times degrade.
MIT 144 installs ★ 114 updated 2026-11-10
npx skills add travisjneuman/.claude --skill performance-engineering -a claude-codeWhat performance-engineering does
This skill guides optimization work across Core Web Vitals (LCP, INP, CLS), bundle splitting, image assets, and caching headers. It enforces a measurement-driven workflow requiring profiling through RUM, Lighthouse, or DevTools before writing any optimization code.
Included guidelines define strict budgets: JavaScript bundles under 200kb, LCP under 2.5s, INP under 200ms, and CLS under 0.1. Load analysis relies on k6 and Artillery, though load tests execute strictly on explicit request.
Reference material sits in standalone files inside the references folder, detailing caching patterns with immutable hashes, stale-while-revalidate strategies, anti-patterns, and review checklists.
What it helps with
- set performance budgets for bundle size and Core Web Vitals
- analyze and split oversized JavaScript bundles
- configure browser and CDN caching headers
- generate load testing scripts with k6 or Artillery
Example prompts for the agent
- Inspect our client bundle and suggest code splitting to meet a 200kb budget
- Set up caching headers for static assets and dynamic endpoints using stale-while-revalidate
- Create a k6 load test script for our checkout API flow
How to install performance-engineering
1Skills CLI (Vercel)
npx skills add travisjneuman/.claude --skill performance-engineering -a claude-code2Manually: put the files in place
~/.claude/skills/performance-engineering/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/performance-engineering/
1Skills CLI (Vercel)
npx skills add travisjneuman/.claude --skill performance-engineering -a codex2Manually: put the files in place
.agents/skills/performance-engineering/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add travisjneuman/.claude --skill performance-engineering -a cursor2Manually: put the files in place
.cursor/skills/performance-engineering/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install travisjneuman/.claude skills/performance-engineering2Manually: put the files in place
.github/skills/performance-engineering/SKILL.md1Skills CLI (Vercel)
npx skills add travisjneuman/.claude --skill performance-engineering -a gemini-cli2Platform CLI
gemini skills install https://github.com/travisjneuman/.claude --path skills/performance-engineeringManually: put the files in place
.devin/skills/performance-engineering/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add travisjneuman/.claude --skill performance-engineering -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about performance-engineering
How to install performance-engineering in Claude Code?
Run npx skills add travisjneuman/.claude --skill performance-engineering -a claude-code in your terminal.
What does performance-engineering do?
It provides guidelines and performance budgets for Core Web Vitals, bundle reduction, CDN caching, and load testing.
What tools are needed for performance-engineering?
It uses profiling inputs from Lighthouse, DevTools, or RUM, and requires k6 or Artillery when executing load tests.
Source text
SKILL.md · MIT
--- name: performance-engineering description: Web Vitals, bundle optimization, CDN, caching, and load analysis. Use when diagnosing slow pages or APIs, reducing bundle size, or designing caching and performance budgets. --- # Performance Engineering ## Overview This skill covers measuring, optimizing, and maintaining web application performance across the full stack. It addresses Core Web Vitals (LCP, INP, CLS), bundle analysis and code splitting, image optimization, browser and server-side caching, CDN configuration, database query performance, and load testing with k6 and Artillery. Use this skill when Core Web Vitals scores are below targets, bundle sizes are growing, page load times are degrading, or when preparing for traffic spikes. Performance work should be measurement-driven: profile first, optimize second, measure the impact third. --- ## Core Principles 1. **Measure before optimizing** - Profile with real data (RUM, Lighthouse, DevTools) before writing any optimization code. Intuition about performance bottlenecks is usually wrong. 2. **Budget everything** - Set performance budgets for bundle size (< 200kb JS), LCP (< 2.5s), INP (< 200ms), CLS (< 0.1). Check changes against them in review. 3. **Optimize the critical path** - Focus on what blocks the user from seeing and interacting with content. Everything else can load later. 4. **Cache aggressively, invalidate precisely** - Use immutable hashes for static assets, short TTLs for dynamic content, and stale-while-revalidate for the best of both worlds. 5. **Reason at realistic scale** - Judge designs against production-like data volumes and traffic, not toy datasets with 10 concurrent users; run load tests only when the user asks for them. --- ## Reference files Detailed material is in separate files so this overview stays short. Read only the file whose topic matches the task; each one is self-contained. Paths are relative to the folder that holds this `SKILL.md` (in the full toolkit install, `~/.claude/skills/performance-engineering/`). | Topic | File | Covers | | --- | --- | --- | | Key Patterns | [`references/key-patterns.md`](references/key-patterns.md) | Pattern 1: Core Web Vitals Monitoring; Pattern 2: Bundle Optimization and Code Splitting; Pattern 3: Image Optimization; Pattern 4: Caching Strategy; Pattern 5: Load Testing with k6 | | Performance Budget Reference | [`references/performance-budget-reference.md`](references/performance-budget-reference.md) | 14 lines | | Anti-Patterns | [`references/anti-patterns.md`](references/anti-patterns.md) | 14 lines | | Checklist | [`references/checklist.md`](references/checklist.md) | 16 lines | ## Related Resources - **Skills:** `monitoring-observability` (performance monitoring), `serverless-development` (cold start optimization) - **Rules:** `docs/reference/checklists/ui-visual-changes.md` (visual performance) - **Rules:** `docs/reference/stacks/react-typescript.md` (React performance patterns)
© travisjneuman · MIT · travisjneuman/.claude