Run a technical website scan

Run NeshatScan to check crawler access, readable content and site signals, with evidence for each finding. The scan is free; email opens the full report.

[method]

What NeshatScan checks, and what it refuses to pretend

There is no composite score. A site is ready when its list of blockers and warnings is empty, not when a made-up number crosses a threshold. Every finding shows the exact line, header or snippet it found. Where a finding rests on a published rule, such as a crawler's documentation or a web standard, it links that source, with the date the source was last read.

Access

Can the answer engines fetch this page at all?

robots.txt behaviour: a 5xx or 429 means “disallow everything” to crawlers; an HTML body means a single-page app is answering for it
Per-crawler verdicts for 14 crawlers and 2 control tokens, read with an RFC 9309 parser that keeps the line number that decided each one
The accidental block: training crawlers blocked by name while the answer crawlers fall through to a wildcard that blocks them too
Cloudflare Content-Signal directives, and hints when an edge provider or a bot challenge sits in front of the site
Indexing directives in headers and meta tags: noindex, nosnippet, max-snippet:0, data-nosnippet
HTTP health: status, redirect chain, HTTPS, time to first byte, size, content type

Understanding

When they fetch it, what do they actually read?

The crawler's-eye view: the text a crawler that runs no JavaScript receives, with a word count and an excerpt
Title, meta description, exactly one H1, page language, canonical URL, Open Graph
Structured data: does it parse, which types are present, does the business entity carry name, url, logo, sameAs and address
FAQ schema must be visible on the page; ratings in schema must be shown on the page
Entity consistency: is the business name spelled the same way in schema, og:site_name and the title

Discovery

Can they find the rest of the site, and what is optional?

Sitemap: found, parses, URL count, host consistency, and the anti-pattern where every lastmod is today
llms.txt: reported as optional, never graded

How findings are labelled

Blocker: Stops an answer engine from fetching or using this page.
Warning: Weakens how well engines understand or trust the page.
Pass: Working as it should.
Info: Your choice, or context worth knowing. Never counted against you.
Optional: Not graded. Reported so you know where things stand.

What it does not do

It does not run JavaScript, on purpose: that is how most AI crawlers read your site. A rendered-vs-raw comparison is planned.
The scan itself does not ask any AI engine whether it names you. The Citation Check does: one buyer question to the assistants for free when the result opens, five questions by email.
It does not fetch more than five URLs per scan, and it never fetches anything that is not a public website. See how the NeshatScan crawler behaves.
The full report opens with your email address, which is stored with the scan so the report link reaches you and Behnood can follow up. Scanned URLs and results are kept so that reports have permanent links.

Questions about checking AI visibility

[01]How do I check my website's AI visibility?

Enter your address above. NeshatScan reads your robots.txt, your page and your sitemap the way a crawler does, without running JavaScript, and shows with evidence whether ChatGPT search, Perplexity, Claude and Google's AI features can reach, read and quote it. It takes a few seconds; the full report opens with your email address.

[02]Can it tell me whether ChatGPT mentions my business?

Yes. After the scan, run the Citation Check: it asks ChatGPT, Gemini, Perplexity and Grok a question a customer of yours would ask, and shows who each one named, you or someone else. It is free, up to three checks a day.

[03]Why doesn't ChatGPT mention my business?

The usual reasons, in order: its crawler cannot reach your pages (a robots.txt rule, a firewall or a security plugin), your pages do not say plainly what you do and where, or other sources do not confirm you. The scan checks the first two; the Citation Check shows who is named instead.

[04]Is there an AI visibility score?

Not here, on purpose: one number hides what to fix. NeshatScan counts blockers, warnings and passes and shows the evidence for each, such as the exact robots.txt line that blocks a crawler, so you know what to change.

[05]Is this AI visibility checker free?

Yes. The scan and the Citation Check are free; the full report opens with your email address. If you want the fixes written for you, or the answers tracked every week, the Fix Pack and Monitoring are on the pricing page.

The steps behind each check are in how to improve brand visibility in AI search engines; to follow the answers week by week, see AI visibility tracking.