Skip to content
RANKCORTEXDemoStart free

Free SEO + AI audit, no signup

See your website the way Google and AI see it

Paste any URL. RankCortex fetches it like ChatGPT, Claude, Perplexity and Google crawlers do, then shows what they can reach, read and understand, and the one fix to make first.

Opens the real audit on rankcortex.com in a new tab.

  • 5 searchOAI-SearchBot, Claude-SearchBot, Googlebot…
  • 3 on-demandChatGPT-User, Claude-User, Perplexity-User
  • 7 trainingGPTBot, ClaudeBot, Google-Extended…
GET https://blog.example/ · no JavaScript

Sample audit of blog.example: 26 checks pass, 2 warn and 2 fail, for a score of 93 out of 100.

30

weighted checks per page

15

AI and search crawlers tested

4

pillars: reach, read, understand, act

24/7

monitoring in the full platform

The platform

Everything your sites need to rank, in one dashboard

The free audit is one page. The platform watches every site you run, every day, and tells you what to fix before traffic drops.

Uptime and health

Every site checked around the clock, with response time, 24h uptime and a clear list of what needs attention.

Sample data
  • shop.example97.2%
  • blog.example100.0%
  • docs.example99.9%
  • app.example99.6%
Open the demo dashboard →

What we check

Four pillars, thirty checks, one score

Each check is weighted by how much it affects whether your page gets found, understood and cited. This is the real scoring model, straight from the product.

How a RankCortex audit turns a URL into one fixFlowchart: your URL is fetched the way crawlers fetch it, as plain HTML with robots.txt, llms.txt and the sitemap. Thirty checks run in four pillars: Reach 30 percent of the score, Read 25, Understand 30 and Act 15. The pillars add up to a score out of 100 and the one fix to make first.GETYour URLany public pageFetch like a crawlerHTML only, no JavaScriptrobots.txt · llms.txt · sitemapReach · 30%Can crawlers fetch the page at all?Read · 25%Is the content in the raw HTML?Understand · 30%Can machines tell what it is about?Act · 15%Can agents use the page?Score / 100pass · warn · failFix firstone changeLEGENDwhat crawlers getpillar, % of scoreHow a RankCortex audit turns a URL into one fixFlowchart, top to bottom: your URL, fetched like a crawler, then four pillars (Reach 30 percent, Read 25, Understand 30, Act 15), then a score out of 100 and the one fix to make first.GETWEIGHTEDYour URLany public pageFetch like a crawlerHTML only, no JavaScriptrobots.txt · llms.txt · sitemapReach · 30%Can crawlers fetch the page at all?Read · 25%Is the content in the raw HTML?Understand · 30%Can machines tell what it is about?Act · 15%Can agents use the page?Score / 100pass · warn · failFix firstone changeLEGENDwhat crawlers getpillar, % of score

Where the points come from

A check's points are its pillar's share times its weight inside the pillar. Real text in the HTML is the single biggest check: 7.8 of 100 points. Open any row for the fix.

Reach Can crawlers fetch the page at all?

  • Reachable by a crawler6.3 points

    Make the page answer 200 to non-browser clients; bot challenges and 403/503 responses hide it from AI crawlers.

  • AI search crawlers allowed6.3 points

    Allow OAI-SearchBot, Claude-SearchBot, PerplexityBot, Googlebot and Bingbot in robots.txt so AI search can cite the page.

  • Indexing and snippet directives5.0 points

    Remove noindex / nosnippet from the meta robots tag and the X-Robots-Tag header.

  • Served over HTTPS3.8 points

    Serve the page over HTTPS and redirect http:// to it.

  • On-demand fetches allowed3.8 points

    Allow ChatGPT-User, Claude-User and Perplexity-User in robots.txt so assistants can open the page when a user asks.

  • Answers quickly2.5 points

    Cut time to first byte below 800 ms with page caching or a CDN; crawlers give up on slow origins.

  • Training crawlers1.3 points

    Always passes: blocking training crawlers is a policy choice, so the report only lists them.

  • Content-Signal declared1.3 points

    Add a Content-Signal line to robots.txt to state how AI systems may use the content.

Read Is the content in the raw HTML?

  • Real text in the HTML7.8 points

    Render the main content on the server; AI crawlers do not run JavaScript and see an empty shell.

  • Page title4.7 points

    Write a descriptive <title> between 10 and 65 characters.

  • Dates the page4.7 points

    Add datePublished and dateModified to the JSON-LD, or a <time datetime> element.

  • Language declared3.1 points

    Add a lang attribute to the <html> element, e.g. <html lang="en">.

  • Meta description3.1 points

    Add a meta description of 50–160 characters that summarizes the page.

  • RSS or Atom feed1.6 points

    Link an RSS or Atom feed with <link rel="alternate" type="application/rss+xml">.

Understand Can machines tell what it is about?

  • Structured data (JSON-LD)4.6 points

    Add a JSON-LD block describing the page (Article, Product, Organization…).

  • One clear H13.5 points

    Use exactly one non-empty <h1> that names the page's topic.

  • Recognized entity types3.5 points

    Use schema.org types that AI systems recognize, such as Article, Product, Organization or FAQPage.

  • Canonical URL3.5 points

    Add an absolute <link rel="canonical" href="https://…">.

  • XML sitemap3.5 points

    Publish a valid XML sitemap and list it in robots.txt.

  • Heading outline2.3 points

    Nest headings without skipping levels (h2 → h3, not h2 → h4).

  • Entity links (sameAs)2.3 points

    Add sameAs links (Wikipedia, LinkedIn, social profiles) to the Organization or Person in the JSON-LD.

  • Author markup2.3 points

    Add an author (Person with name and url) to the Article JSON-LD.

  • Open Graph tags2.3 points

    Add og:title, og:description and og:image meta tags.

  • llms.txt2.3 points

    Publish /llms.txt (llmstxt.org format) listing the site's key pages.

Act Can agents use the page?

  • HTML landmarks3.5 points

    Wrap content in <main> and navigation in <header> or <nav>.

  • Images have alt text3.5 points

    Add alt text to images (alt="" for purely decorative ones).

  • Buttons have names2.3 points

    Give every button visible text, an aria-label or a title.

  • Links have text2.3 points

    Give every link text or an aria-label; icon-only links need a label.

  • Form fields have labels2.3 points

    Associate a <label> or aria-label with every input, select and textarea.

  • WebMCP tools1.2 points

    Optional: passes when the page declares WebMCP tools, skipped when it doesn't.

How it works

From URL to fix list in three steps

  1. 1

    Enter your URL

    Any public page: home page, article, product. No account, no install, no tracking script.

  2. 2

    We crawl it like a bot

    Plain HTML with no JavaScript, plus robots.txt, llms.txt and your sitemap, exactly as AI and search crawlers fetch them.

  3. 3

    Fix what matters first

    A score out of 100, one fix to make first, and a pass, warn or fail for every check with a clear reason.

Pricing

Start free, go Pro when you need more

Sign in with Google or GitHub. No card for the free audit.

Free

$0

One audit to see where you stand

  • 1 full SEO and AI audit
  • Score, fix-first and all 30 checks
  • Crawler access for 15 bots
Run my free audit

Most popular

Pro

$1 for 7 days

Then $9.99/month. Cancel any time.

  • Unlimited audits, every page you own
  • Re-run and share any result
  • 10-minute uptime, Search Console, indexing and scheduled checks
  • Secure checkout by Stripe
Start 7-day trial for $1

Billed by Stripe on rankcortex.com. This demo takes no payments.

FAQ

Questions, answered

Find out what Google and AI miss on your site

Free, instant, no signup. Takes about ten seconds.

Check my website