Skip to content
Solidus kodasAudit

Is your website accessible to AI search, and how well is it protected?

Check for free whether your website's content is technically accessible to AI search, which security weaknesses and outdated software it reveals publicly, and find out what is worth fixing.

Free, no registration. We only check publicly accessible pages; an audit usually takes a few seconds, and up to 120 seconds for a slow website.

How the score is built: the width of each bar matches the group's weight

  • 25%Access
  • 22%Content readability
  • 14%Discoverability
  • 11%Semantic clarity
  • 18%Security
  • 10%Speed

How we calculate the score and what the audit does not prove is described in the methodology.

How it works

  1. You enter an address

    A domain or the URL of a specific public page. No registration needed.

  2. We fetch the pages safely

    We read robots.txt, the sitemap and a small sample of pages. We do not log in, fill in forms or bypass protection.

  3. We run 33 checks

    Each check ends in one of these states: OK, warning, problem, could not be determined or not applicable.

  4. You get a report with evidence

    You see what we found, what it means, what to do and how we checked it – with the URL we checked and an excerpt of the response.

What we check

The overall score is made up of the groups below. The security group covers encryption, security headers, cookies, integrity of external files, e-mail protection, the software versions the site discloses and the vulnerabilities known for them. Additional agent features (llms.txt, Markdown alternatives, signs of WebMCP) and analytics tags (GA, GTM, Meta Pixel) are shown for information only and do not count towards the scores.

Access25% of the score

Whether the page can be reached and whether crawlers are allowed to read and index it.

Content readability22% of the score

Whether the main content is in the HTML returned by the server and whether the topic of the page is clear.

Discoverability14% of the score

Whether crawlers can find the site's pages through the sitemap, canonical links and internal links.

Semantic clarity11% of the score

Whether structured data and language signals help to understand the content unambiguously.

Security18% of the score

Encryption, security headers, cookies, the software versions the site discloses and the vulnerabilities known for them. Part of the overall score; the group score is also shown on its own as the security score.

Speed10% of the score

How quickly the server responds and what the HTML makes the browser wait for: response time, compression, render-blocking scripts, images. This is not a Core Web Vitals measurement in a browser.

Example of a finding

An illustration of what a single finding looks like in the report. This is not data from a real website.

Meta robots and X-Robots-Tag

What we found
The page's response contains the header “X-Robots-Tag: noindex”.
What it means
The page itself asks not to be indexed, even though robots.txt allows access.
What to do
If the page should be findable, remove this header in the server settings.
How we checked
We fetched the page over HTTPS and read the response headers and meta tags.

What the audit shows and what it does not prove

What we can reasonably establish

  • Whether the page and its text can be fetched from our infrastructure.
  • What robots.txt declares to specific crawlers – separately for search, model training and user-initiated access.
  • Whether the page declares restrictions on indexing and snippets.
  • Whether content, internal links and structured data are reachable in the sample we checked.
  • Whether additional files intended for AI and signs of integrations have been published.
  • How encryption, security headers, cookies and protection against e-mail spoofing are configured.
  • Which software and versions the site discloses publicly, and which vulnerabilities are known for those versions.

What an audit like this alone does not prove

  • That ChatGPT or another system has already indexed the site.
  • That the site will be cited or recommended in answers.
  • That our successful request means a real crawler's request from its own network succeeds too.
  • That the few pages we checked represent every product, category or language version.
  • That the site is secure or vulnerable: we only assess publicly visible signs; we do not look for weaknesses or try to exploit them.

Frequently asked questions

Does the audit show that ChatGPT or another assistant already sees my website?

No. The audit checks the technical conditions: whether the page and its text can be fetched, what robots.txt declares to crawlers, and whether indexing is prohibited. Whether a particular AI system has already indexed the site, or will cite it, cannot be established from a check like this.

Is an llms.txt file required?

No. llms.txt, Markdown alternatives and WebMCP are additional options for agents. The report shows them in a separate informational section and they do not count towards the score; their absence does not mean that AI search cannot see the site.

Do you check the site's security as well?

Yes. The security group, which is part of the overall score, rates what is publicly visible: encryption, security headers, cookie attributes, integrity of external files, protection against e-mail spoofing, the software versions the site discloses and the vulnerabilities known for them. We do not look for hidden files, scan ports or try to exploit vulnerabilities, so this is not a penetration test.

How many pages do you check?

The quick audit checks the page you entered, robots.txt, the sitemap and up to 5 selected pages from the same domain. Every report shows the exact list of addresses checked, because the findings apply to that sample, not to the whole site.

What does the score mean?

The overall score is a rough guide made up of these groups: access (25%), content readability (22%), discoverability (14%), semantic clarity (11%), security (18%) and speed (10%). The security score is the score of the security group, shown on its own as well. This is a choice made in our methodology, not a model of how AI systems rank sites. Critical blockers are shown separately so that a high score does not hide them.

Why do I sometimes see “Insufficient data” instead of a score?

If essential data could not be obtained – for example, the site did not respond in time or protection stopped our request – we do not show an overall score. Such checks are marked “Could not be determined” rather than treated as a fault of the site.

Do I need to register or leave my e-mail address?

No. The whole report is visible without registration and without an e-mail address. You only leave your e-mail if you would like a quote for fixing the issues found.

Who can see my report?

The report is available through an unguessable link. Anyone you pass the link to will see it, so share it with care. Report pages are not indexed by search engines.