Website SEO Audit Crawler - Broken Links, Meta & Page Speed

Crawl an entire website and audit every page for technical SEO problems: broken links and images, missing or duplicate titles and meta descriptions, missing H1s, thin and duplicate content, redirect chains, canonical errors, slow pages and Core Web Vitals. Each page gets an on-page score from 0 to 100 and a plain-English list of errors, warnings and notices, worst pages first. It's a pay-per-page alternative to Screaming Frog, Sitebulb and the Ahrefs or Semrush site audit, with no install and no subscription.

What you get

One row per page, sorted worst first:

• url, statusCode, redirectTo and onpageScore (0-100)
• errors, warnings and notices: readable issue lists such as "Broken links (3)", "Missing meta description", "Duplicate title", "Redirect chain", "Images missing alt text" and "Slow page load", with errorCount, warningCount and noticeCount
• brokenLinks and brokenLinkCount: the broken URLs on the page, with their status codes
• title, titleLength, description, descriptionLength, h1, h1Count and canonical
• isIndexable and nonIndexableReason (noindex, robots.txt, canonicalised…)
• wordCount, readabilityScore, clickDepth, internalLinks, externalLinks and inboundLinks
• duplicateTitle, duplicateDescription and duplicateContent
• pageSizeKb, loadTimeMs, serverResponseMs, plus timeToInteractiveMs, largestContentfulPaintMs and cumulativeLayoutShift in the browser-rendered modes

The SUMMARY record in the Key-value store tab has the site-wide score, total errors, warnings and notices, every issue type with the number of pages it affects, and server, CMS and SSL certificate details.

The Dataset tab has views for issues, content and meta, broken links, and speed. Export to CSV, Excel, JSON or Google Sheets.

Use cases

• Technical SEO audits: find everything holding a site back in one crawl
• Broken link checking: list every broken internal and external link and the page it's on
• Agency client onboarding: produce an audit for a prospect or new client in minutes
• Site migrations and redesigns: crawl before and after launch to catch broken pages and redirect chains
• Content cleanup: find thin, duplicate and non-indexable pages
• Page speed and Core Web Vitals: find the slowest pages across the whole site, not just the home page

How to use

1. Enter the Website URL, e.g. example.com or https://www.example.com/blog/ to start from a section.
2. Set Max pages (default 100, up to 10,000).
3. Pick a Crawl mode:
   • Standard (HTML): fast and cheapest, right for most sites
   • JavaScript rendering: for React, Vue, Angular and other sites whose content loads with JavaScript
   • JavaScript + Core Web Vitals: also measures LCP, CLS and time to interactive in a real browser
4. Optionally turn on Include subdomains, and keep Use sitemap.xml on to find orphan pages.
5. Run the actor. A 100-page standard crawl usually takes 3-5 minutes.

Output example

{
  "url": "https://books.toscrape.com/",
  "statusCode": 200,
  "onpageScore": 91.3,
  "errorCount": 0,
  "warningCount": 4,
  "noticeCount": 3,
  "errors": [],
  "warnings": ["Missing meta description", "Thin content", "Duplicate title", "Duplicate content"],
  "notices": ["Render-blocking resources", "Low text-to-HTML ratio", "Missing charset meta tag"],
  "title": "All products | Books to Scrape - Sandbox",
  "titleLength": 40,
  "description": null,
  "h1": "All products",
  "isIndexable": true,
  "wordCount": 180,
  "clickDepth": 0,
  "internalLinks": 73,
  "brokenLinks": null,
  "brokenLinkCount": 0,
  "pageSizeKb": 50.1,
  "loadTimeMs": 328
}

Tips

• Start with the errors: broken pages, broken links and missing titles hurt most. Notices are minor improvements.
• If a JavaScript-heavy site shows almost no content or links, rerun it in JavaScript rendering mode.
• Schedule a weekly run to catch new broken links before your visitors and search engines do.

Pricing

You pay per page audited. JavaScript rendering and Core Web Vitals pages cost more because each page is loaded in a real browser. There's no subscription, and the crawl stops automatically at your run's maximum charge.