---
id: wiki-2026-0508-google-lighthouse
title: Google Lighthouse
category: 10_Wiki/Topics
status: verified
canonical_id: self
aliases: [Lighthouse, Web Vitals Audit]
duplicate_of: none
source_trust_level: A
confidence_score: 0.9
verification_status: applied
tags: [performance, web-vitals, audit, ci]
raw_sources: []
last_reinforced: 2026-05-10
github_commit: pending
tech_stack:
  language: JavaScript
  framework: Lighthouse 12+ / lighthouse-ci
---

# Google Lighthouse

## 매 한 줄
> **"매 web page 의 performance, accessibility, best-practices, SEO, PWA 를 자동 audit"**. Chrome DevTools 내장, CLI, Node API, lighthouse-ci 의 form. 2026 현재 v12+ 가 INP (Interaction to Next Paint) 의 primary metric, LCP/CLS/INP 의 Core Web Vitals 와 align. 매 PR 마다 budget 검증 의 standard practice.

## 매 핵심

### 매 5 categories
- **Performance**: LCP, INP, CLS, TBT, FCP, Speed Index → 0-100 score.
- **Accessibility**: axe-core 기반 a11y 검사.
- **Best Practices**: HTTPS, secure headers, deprecated API.
- **SEO**: meta tags, mobile viewport, structured data.
- **PWA** (deprecated 2024 후 separate plugin).

### 매 Performance 의 weight (v12)
- LCP 25%, TBT 30%, CLS 25%, FCP 10%, SI 10%.
- INP 의 field metric 만 — lab 에서 measure X (Total Blocking Time 으로 proxy).

### 매 응용
1. Local audit (DevTools).
2. CI gate (lighthouse-ci, budgets.json).
3. Field monitoring 보조 (CrUX + RUM 와 결합).
4. PageSpeed Insights backend.

## 💻 패턴

### CLI run
```bash
npm i -g lighthouse
lighthouse https://example.com --output=html --output=json \
  --output-path=./report --chrome-flags="--headless"
```

### Node programmatic
```javascript
import lighthouse from 'lighthouse'
import * as chromeLauncher from 'chrome-launcher'

const chrome = await chromeLauncher.launch({ chromeFlags: ['--headless'] })
const result = await lighthouse('https://example.com', {
  port: chrome.port,
  output: 'json',
  onlyCategories: ['performance'],
})
console.log(result.lhr.categories.performance.score)
await chrome.kill()
```

### Lighthouse CI in GitHub Actions
```yaml
# .github/workflows/lhci.yml
name: lhci
on: [pull_request]
jobs:
  lhci:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with: { node-version: 20 }
      - run: npm ci && npm run build
      - run: npm i -g @lhci/cli@0.14.x
      - run: lhci autorun
        env:
          LHCI_GITHUB_APP_TOKEN: ${{ secrets.LHCI_GITHUB_APP_TOKEN }}
```

### lighthouserc.js with budgets
```javascript
module.exports = {
  ci: {
    collect: {
      url: ['http://localhost:3000/', 'http://localhost:3000/about'],
      numberOfRuns: 3,
      startServerCommand: 'npm start',
    },
    assert: {
      assertions: {
        'categories:performance': ['error', { minScore: 0.9 }],
        'largest-contentful-paint': ['error', { maxNumericValue: 2500 }],
        'cumulative-layout-shift': ['error', { maxNumericValue: 0.1 }],
        'total-blocking-time': ['warn', { maxNumericValue: 200 }],
      },
    },
    upload: { target: 'temporary-public-storage' },
  },
}
```

### User flow (multi-step)
```javascript
import { startFlow } from 'lighthouse'
import puppeteer from 'puppeteer'

const browser = await puppeteer.launch()
const page = await browser.newPage()
const flow = await startFlow(page, { name: 'Checkout flow' })
await flow.navigate('https://shop.example.com')
await flow.startTimespan({ name: 'Add to cart' })
await page.click('#add')
await flow.endTimespan()
await flow.snapshot({ name: 'Cart loaded' })
const report = await flow.generateReport()
```

## 매 결정 기준
| 상황 | Approach |
|---|---|
| Local optimization | DevTools Lighthouse panel |
| PR budget gate | lighthouse-ci + assertions |
| Multi-page audit | lhci with multiple URLs |
| SPA route transitions | User flows API |
| Real-user metrics | CrUX / web-vitals JS (Lighthouse 의 X) |

**기본값**: 매 lhci + budgets.json + GitHub Action + 3 runs median.

## 🔗 Graph
- 부모: [[Performance_Profiling_and_Memory|Performance Profiling]]
- 응용: [[CI_CD_Pipeline|CI_CD Pipeline]] · [[Continuous Integration (CI)]]
- Adjacent: [[Core Web Vitals Optimization (INP, LCP 개선)|Cumulative Layout Shift (CLS)]] · [[SPA 라우트 전환 성능 최적화]]

## 🤖 LLM 활용
**언제**: pre-deploy quality gate, regression catch, web vitals tracking.
**언제 X**: 매 real-user perf 측정 (lab condition 일 뿐 — RUM 사용). Native app, server-only API.

## ❌ 안티패턴
- **단일 run score 의 의존**: variance 가 큼 — 매 median of 3-5 runs.
- **Lab score = field score 의 가정**: throttling profile 이 user 와 다름. 매 CrUX 와 cross-check.
- **High score 만 의 chase**: 100 score ≠ good UX. 매 INP/LCP 의 actual budget 의 focus.
- **CI 에 emulated mobile 만 의 test**: desktop user 도 audit.

## 🧪 검증 / 중복
- Verified (web.dev/lighthouse, lighthouse-ci docs, Lighthouse 12 release notes 2025).
- 신뢰도 A.

## 🕓 Changelog
| 날짜 | 변경 |
|---|---|
| 2026-05-08 | Phase 1 |
| 2026-05-10 | Manual cleanup — Lighthouse 12 + CI integration patterns |