BlinkHub
Free score · full report with a free account

AI auto-diagnoses your SEO
and writes the fix code for you.

Paste any URL. We score your site on traditional SEO + AI-search readiness, then generate paste-ready fix code, written by our AI.

🔓 Free score💳 No credit card🔒 No data storedEarly access

📋 WHAT YOU GET

Before → After: what the AI fix code looks like

A real example. Paste the snippet into your <head> and your score jumps in minutes.

BEFORE — what most sites have38/100
<title>About</title>
<!-- no meta description -->
<!-- no Open Graph tags -->
<!-- no JSON-LD structured data -->
<!-- no FAQ schema -->
AFTER — AI-generated paste-ready snippet91/100
<title>Acme — AI-Powered Customer Support for SaaS (2026)</title>
<meta name="description" content="Acme is the AI customer-support agent built for SaaS teams. Auto-replies, ticket triage, and human-handoff in one console. Free 14-day trial.">
<link rel="canonical" href="https://acme.io/">
<meta property="og:type" content="website">
<meta property="og:title" content="Acme — AI Customer Support for SaaS">
<meta property="og:description" content="AI-powered ticket triage with seamless human handoff.">
<script type="application/ld+json">{"@context":"https://schema.org","@type":"SoftwareApplication","name":"Acme","applicationCategory":"BusinessApplication","offers":{"@type":"Offer","price":"0","priceCurrency":"USD"}}</script>
38AI fix91(+53)

Score in sample jumped from 38 → 91. Your result depends on your site.

💳 GET THE FIX CODE

Sign up & buy credits to unlock AI fix code

Each SEO audit fix costs 100 credits. Pick a pack — the bigger the pack, the bigger the bonus.

🔒 PayPal (global) · Bank transfer (Korea)

Why we score AI search separately

Search is splitting

40% of Gen Z now starts queries on AI search engines, not just Google. Citation signals diverge from traditional ranking.

Schema matters more

LLMs cite structured data (FAQ, Article, llms.txt) far more often than unstructured pages. Our audit checks both.

Fix code, not advice

Most auditors hand you a checklist. We hand you a paste-ready <head> snippet generated by our AI — apply in 30 seconds.

Built into BlinkHub

BlinkHub is an AI publishing + algorithm-automation suite. This audit tool is one of three AI tools: SEO, social strategy, and upload-time optimization — all on a unified credit system.

See pricing →

What this audit looks at

Paste a URL and the page is fetched and read the way a crawler reads it. You get two separate scores — one for traditional search, one for whether an AI assistant could quote the page — plus the list of what is missing.

Indexability

Canonical URL
Whether the page declares one, and whether it points at itself or somewhere else.
Robots meta
noindex / nofollow directives that quietly keep a page out of search.
Language attribute
The lang value, which decides how a page is matched to a locale.
Viewport
Missing viewport is why a page fails mobile evaluation outright.

What appears in results

Title
Present, and its length. Duplicated brand names get rewritten by Google.
Meta description
Present, and its length — too short leaves Google to invent one.
H1
How many exist. Zero or several both confuse the topic signal.
H2 structure
Whether the page is sectioned at all, which decides how quotable it is.
Open Graph title and image
What a link looks like when shared.

AI search readiness

Structured data
Whether JSON-LD exists and which @type values are declared.
FAQ schema
The format assistants quote most often, because question and answer are already paired.
Article schema
Signals that this is a piece with an author and a date rather than a landing page.
Author and date
Whether the page states who wrote it and when — assistants weigh both.
llms.txt
Whether the site publishes a plain-language summary of what it is.

Accessibility of content

Images without alt text
Counted, because an assistant reading the page sees nothing where alt is missing.

Two scores, because they are two different jobs

Search score

Can this page be found and shown in a list of results? Canonicals, robots directives, title and description, heading structure, mobile viewport, image alt text.

AI readiness score

Could an assistant quote this page in an answer? Structured data and its types, FAQ and Article schema, a stated author and date, and whether the site publishes llms.txt.

A page often passes one and fails the other. A well-optimised marketing page with no schema, no author and no date scores well on search and poorly on AI readiness — it ranks, but assistants have little reason to cite it.

The problems we see most often

  1. 01A title template that doubles the brand nameA layout appends “| Brand” while each page title already ends in it. Google rewrites such titles, and pages can be crawled and then left unindexed.
  2. 02hreflang declared for languages that do not existPointing hreflang at a URL that returns 404 is worse than declaring nothing.
  3. 03Structured data that disagrees with the pageFAQ schema for questions that are not visible on the page is a policy problem, not a shortcut.
  4. 04A sitemap full of dead URLsEvery 404 submitted spends crawl budget that a live page could have had.
  5. 05Content that only exists after JavaScript runsIf the text is not in the served HTML, assume the crawler may not see it.
  6. 06A stale llms.txtPrices and features that changed months ago will be repeated back to users as current.

Frequently asked

What does the SEO audit actually fix?

The audit itself reports problems; the fix step writes the code. For a missing description it drafts one to length. For absent structured data it produces the JSON-LD block with your page's real title, author and date filled in. For a duplicated title it proposes a rewritten one. You paste the output into your page — nothing is changed on your site by us.

How is an AI search audit different from an SEO audit?

An SEO audit asks whether a page can rank in a list of links. An AI search audit asks whether an assistant can quote it. Those need different things. Ranking rewards canonicals, titles and speed. Being quoted rewards sentences that answer a question directly, numbers with a stated basis, question-shaped headings, and schema that matches what is visible. This tool scores both separately, so a page can pass one and fail the other.

Do I need llms.txt?

It is not required and not a ratified standard. It is a plain-text summary at your site root describing what the site is. Assistants may use it when summarising you; there is no guarantee any given one reads it. It costs almost nothing to publish, so the audit flags its absence — but a stale llms.txt is worse than none, because it teaches assistants facts that are no longer true.

Should I block AI crawlers in robots.txt?

Only if you do not want to be quoted. Blocking GPTBot, ClaudeBot, PerplexityBot or Google-Extended removes your pages from those systems' answers. If visibility in AI answers is the goal, blocking them works directly against it. Decide the goal first; the technical choice follows.

How often should I run it?

After any change to templates, metadata or structured data, and then on a monthly rhythm. Most regressions are introduced by a template edit rather than by content — a title template that appends a brand name that the page title already contains, for example, will silently affect every page at once.

Is it really free, and do I need an account?

Running an audit on a URL is free and needs no account. The AI-written fix code is a paid step, charged in credits, because it calls a model each time. You can see the full score and issue list before deciding whether you want the fixes.

Related