Lighthouse Checker

Run structured Lighthouse audits for one or more URLs and get API-ready results with normalized scores, key metrics, and the top failed audits for each page.

Why this actor exists

Most site audit tools either return too much raw Lighthouse data or bundle Lighthouse into a broader crawler. Lighthouse Checker stays narrow:

• batch Lighthouse audits for one or more URLs
• mobile or desktop presets
• normalized scores for performance, accessibility, best practices, and SEO
• key metrics like FCP, LCP, Speed Index, TBT, and CLS
• top failed audits with short recommendation text
• clean JSON output for automations, QA workflows, and agency reporting backends

Best fit for:

• developers and technical SEO operators
• agencies auditing multiple landing pages
• QA teams that want repeatable page-quality checks
• automations that need structured Lighthouse results, not raw report sprawl

What you get

Each result returns:

• target URL and final URL
• device preset used
• normalized 0-100 scores
• key Lighthouse performance metrics
• top failed audits ranked by severity
• concise recommendation text
• clean per-URL error objects when an audit fails or times out

Input

Minimal run

{
  "startUrls": ["https://example.com"]
}

Batch run

{
  "startUrls": [
    "https://example.com",
    "https://apify.com"
  ],
  "device": "desktop",
  "maxUrls": 10,
  "requestTimeoutSecs": 45,
  "topAudits": 5
}

Example output

{
  "summary": {
    "requestedUrlCount": 2,
    "auditedUrlCount": 2,
    "successCount": 2,
    "errorCount": 0,
    "device": "desktop",
    "timeoutSecs": 45,
    "topAuditsPerUrl": 5
  },
  "results": [
    {
      "url": "https://example.com/",
      "device": "desktop",
      "status": "ok",
      "scores": {
        "performance": 99,
        "accessibility": 100,
        "bestPractices": 93,
        "seo": 80
      },
      "metrics": {
        "firstContentfulPaintMs": 777.52,
        "largestContentfulPaintMs": 777.52,
        "speedIndexMs": 777.52,
        "totalBlockingTimeMs": 0,
        "cumulativeLayoutShift": 0
      },
      "topFailedAuditCount": 4,
      "topFailedAudits": [
        {
          "id": "meta-description",
          "title": "Document does not have a meta description",
          "score": 0,
          "displayValue": null,
          "recommendation": "Write a concise meta description that reflects the page content."
        }
      ],
      "error": null
    }
  ]
}

Full example files live in:

• examples/basic-input.json
• examples/agency-batch-input.json
• examples/sample-output.json
• store-assets/pricing.md
• store-assets/screenshots.md
• store-assets/positioning.md

Pricing recommendation

Start with simple usage-based pricing in the Apify Store:

• free: small trial usage so buyers can validate output shape
• paid: usage-based per compute unit / per run through Apify standard billing
• positioning: cheaper than manual audits, faster than custom Lighthouse scripting

Detailed pricing notes are in store-assets/pricing.md.

Positioning

Positioning line: Run batch Lighthouse audits and get clean JSON summaries you can plug into automations, dashboards, or client workflows.

Not positioned as:

• a full crawler
• a dashboard product
• an authenticated session tester
• a PDF reporting tool

That tighter scope keeps the actor easier to understand, cheaper to try, and more store-friendly.

Categories and keywords

Recommended Apify Store categories:

• SEO
• Developer tools
• Monitoring and alerting

Core keywords:

• lighthouse
• pagespeed
• performance
• accessibility
• seo
• website audit
• qa
• developer tools

Limits in v1

Deliberately not included in this version:

• dashboards or historical trends
• authenticated or cookie-injected audits
• custom throttling profiles
• PDF or HTML report exports
• screenshots, filmstrips, or trace artifacts in run output
• field-data enrichment from external APIs
• multi-run comparisons or diffing

Local verification

cd ~/clawd/income-engine/streams/lighthouse-checker
npm run health
npm run test:local
node src/main.js

Author

Built by Janice / literal_jacktree.