🚨 VitalsBeacon - Core Web Vitals API

Lighthouse Performance Audits for Any Website

🚀 Quick Test

Test Google.com (Quick Mode)

Test Example.com (Full Audit)

Test Stripe.com (Mobile)

📖 API Usage

Basic Audit:

GET /audit?url=https://example.com

Returns all categories: Performance, Accessibility, Best Practices, SEO

Quick Mode (Performance Only):

GET /audit?url=https://example.com&quick=true

Faster! Only runs performance audit (~15-20s instead of 30-40s)

Mobile vs Desktop:

GET /audit?url=https://example.com&strategy=mobile

GET /audit?url=https://example.com&strategy=desktop

Test mobile or desktop experience (default: desktop). Desktop uses Lighthouse's own desktop throttling (desktopDense4G, no CPU slowdown), so desktop scores line up with PageSpeed Insights and Chrome DevTools. Mobile uses Lighthouse's mobile calibration.

Custom Categories:

GET /audit?url=https://example.com&categories=performance,seo

Available Categories:

• performance - Core Web Vitals & speed metrics
• accessibility - A11y compliance
• best-practices - Web best practices
• seo - SEO optimization

Repeat Runs (median of 3):

GET /audit?url=https://example.com&runs=3

Runs the performance audit three times and reports the MEDIAN score and that run's metrics, plus a variance object ({ runs, performanceScores, spread }) - steadier numbers on noisy sites. Allowed values: 1 (default) or 3. Heads up: runs=3 takes about 3x as long and counts as 3 requests against your rate limit.

Top Opportunities:

Whenever performance is audited, the response includes an opportunities array - Lighthouse's top 3 fixes by estimated savings (omitted when there are none):

"opportunities": [
  {
    "id": "render-blocking-resources",
    "title": "Eliminate render-blocking resources",
    "savings": "1.2 s",
    "savingsMs": 1230
  }
]

Field Selection:

GET /audit?url=https://example.com&fields=scores

Return only what you need. Comma-separated from scores, coreWebVitals, opportunities, variance. The response always keeps url, timestamp, strategy, cached (and age when cached). fields=scores is the compact badge/CI shape.

Response Format:

{
  "url": "https://example.com",
  "timestamp": "2025-10-17T12:00:00.000Z",
  "strategy": "desktop",
  "scores": {
    "performance": 95,
    "accessibility": 88,
    "bestPractices": 92,
    "seo": 90
  },
  "coreWebVitals": {
    "largestContentfulPaint": {
      "value": "1.2 s",
      "score": 100,
      "rating": "good"
    },
    ...
  },
  "cached": false
}

Cached responses also include an age field: seconds since the result was generated.

Response Headers:

• X-Cache - HIT or MISS
• X-RateLimit-Remaining - audits left in the current window
• X-RateLimit-Reset - Unix time (seconds) when the window resets

Error Format:

All errors share one shape with a stable machine-readable code:

{
  "error": "Rate limit exceeded",
  "code": "rate_limited",
  "hint": "Try again in 42 minutes (limit 50/hour)."
}

Codes: rate_limited, queue_full, timeout, unreachable, invalid_url, invalid_categories, invalid_runs, invalid_fields, internal.

⚡ Performance Tips:

• Results are cached for 24 hours - second request is instant!
• Use quick=true for faster audits
• Mobile and desktop are cached separately
• Audits run one at a time for accurate scores; up to 10 can wait in the queue
• Rate limit: 50 audits per hour per IP

Other Endpoints:

• /health - Service health check
• /status - Queue & memory stats
• /cache-stats - Redis cache info

Built with ❤️ using Lighthouse, Hono, and Bun