Skip to content
GeoProbeCheck your site

YOUR WEBSITE, THROUGH A DIFFERENT LENS

Built for people.
Ready for AI?

Find what’s stopping AI from reading your website. Get the evidence, understand the issues, and know exactly what to fix.

Free to useNo signupUsually 15 seconds

Scans public web content. Reports are public and may be indexed.

Take a look at an example report
THE OUTPUT, NOT JUST A NUMBERFIG. 01
GeoProbe
Illustrative example

AI readiness overview

acme-saas.com

8 checks completed · 4 need attention

58/100

Room to improve

Start here

01 / 04

Your robots.txt blocks AI crawlers

A rule prevents the checked crawlers from accessing your pages.

robots.txt EVIDENCE
User-agent: *
Disallow: /
Review the fix ~2 min
Structured dataPassed
6 more checks, with evidence
RULE-BASED CHECKS. EXPLAINED RESULTS.

A small change in your site. A clearer signal to AI.

8 transparent checks

Every weight, out in the open.

Evidence behind every issue

See what we actually found.

A practical next step

Go from diagnosis to a fix.

01 / FROM QUESTION TO CLARITY

Less guesswork.
A clear way forward.

You don’t need another dashboard to watch.
You need to know what to do next.

01

Give us your URL

We inspect your public pages and crawler instructions. No installation or account needed.

YOUR WEBSITE

02

See what’s getting in the way

Read a prioritised report with the evidence behind each finding. Every check has an explanation.

YOUR DIAGNOSIS

03

Make the next right change

Follow the fix, review your generated llms.txt, then rescan to check your progress.

YOUR NEXT STEP

02 / EVIDENCE COMES FIRST

A score is a starting point.
The proof is the point.

Every finding should answer three questions: what did you find, why does it matter, and what can I do about it?

Explore the example report
THE GEOPROBE APPROACH
01

Show the observation

The file, the rule, the text we found.

02

Explain the interpretation

A clear reason for the result, with its limits.

03

Give a useful next step

A specific fix you can review and apply.

03 / THE METHODOLOGY

Eight checks.
Nothing hidden.

A practical view of access, content and context. Here is exactly how each check contributes to the score.

Our rubric. Not a ranking prediction.

Weights reflect GeoProbe’s diagnostic priorities. They do not measure the likelihood of an AI recommendation.

Understand the limitations
CHECK / SIGNALWEIGHT
01

AI crawler access

Permissions in robots.txt for the AI agents we check.

20pts
02

Content in raw HTML

How much text is available without running JavaScript.

20pts
03

Structured data

Organization, website and page-type JSON-LD.

15pts
04

External presence

Mentions across available third-party sources.

15pts
05

llms.txt

Presence and basic format of this optional content map.

10pts
06

Heading & introduction

An H1 and a concise introductory paragraph.

10pts
07

Page metadata

Title, description, canonical and Open Graph tags.

5pts
08

Content freshness

Recent modification dates in page or sitemap metadata.

5pts
Total possible weight100 pts

Pass earns full weight; warning earns half. Skipped checks are excluded and the remaining weights are normalised to 100.

A LITTLE MORE CLARITY

Good questions.
Straight answers.

Something else on your mind?

hello@geoprobe.ai
What does my score actually tell me?

Your score summarises eight rule-based checks of technical readiness and selected external signals. It is a diagnostic guide, not a measurement of how often AI recommends you. The weights are GeoProbe’s own rubric, not an AI provider’s ranking formula.

Will a higher score get me recommended by ChatGPT?

There is no guarantee. Fixing an access issue can make your content easier to retrieve, but recommendations also depend on relevance, source selection and the AI system. Search access and permission to use content for training are separate choices.

What do you access, and is my report public?

We fetch public website content, including your homepage, robots.txt, llms.txt, llms-full.txt and sitemaps. Generating llms.txt may fetch linked public pages, and the presence check queries external sources. We do not execute your JavaScript. Reports are public, shareable and may be indexed by search engines.

Is it really free?

Yes. No account, email or credit card is needed. There is a limit of five scans per hour per IP. Recent reports may be reused for one hour; Rescan requests a fresh check.

Do I need an llms.txt file?

llms.txt is a proposed convention for giving AI tools a concise content map. It is optional and does not guarantee indexing or citations. We can generate a starting point from your site for you to review before publishing.

Can I check my site again after making changes?

Yes. Open your report and select Rescan to fetch the site again. Compare the evidence and individual checks to confirm what changed.

A CLEARER PICTURE STARTS HERE

See your site in a new light.

Your next improvement could be one scan away.

Check my website