We use necessary cookies to run the shop and your account. Other cookies, such as statistics, are only used if you allow them. You can change your choice at any time under “Cookie settings”. Cookie policy
Cookie settings
Choose which cookies we may use. Necessary cookies are always on. You can change or withdraw your choice at any time with the “Cookie settings” link. Cookie policy
Necessary
Always on
Needed for the website to work and to keep it secure, for example the shopping cart, logging in and remembering your cookie choice. They cannot be turned off.
5 services
Cookie choice · GetEasySoft Remembers your cookie choices, their version and your consent ID. Cookies: esk_consent · 6 months
WordPress · GetEasySoft Logging in to your account and security. Cookies: wordpress_logged_in_*, wordpress_sec_*, wordpress_test_cookie · session or up to 14 days
WooCommerce · GetEasySoft Shopping cart and checkout. Cookies: woocommerce_cart_hash, woocommerce_items_in_cart, wp_woocommerce_session_* · up to 2 days
Stripe · Stripe Payments Europe Ltd. Fraud prevention during card payments. Cookies: __stripe_mid, __stripe_sid · up to 1 year
CloudflareProtection against bots Cookies: __cf_bm, cf_clearance · up to 30 minutes / 1 year
Remember your settings and enable extra features such as embedded maps.
Not used on this website at the moment.
Help us understand how visitors use the website, so we can improve it. The data is collected in aggregate.
Not used on this website at the moment.
Used to measure advertising and show relevant ads on other websites, for example Google and Meta, and for embedded videos.
Check whether ChatGPT, Claude, Gemini, Perplexity and other AI assistants can access, read and understand your website. Free, with plain-language results and the official sources behind every rule.
Checked against the official documentation of the AI vendors, RFC 9309 and schema.org.
Free, no account needed
Only public information: nothing is changed on your site
Results in plain language, with the evidence
What we check
What does the AI Readiness Checker check?
The checker answers three questions about your website: may and can AI crawlers access it, can they read the content, and can they understand what it is about. Every rule comes from an official source: the documentation of the AI vendors, the robots.txt standard RFC 9309, schema.org and Google Search Central.
Area
What we check
Included in
Access (35 points)
What robots.txt allows each of the 27 official AI crawlers, on the home page and the pages linked from it; a test with each crawler’s User-Agent against your firewall (for example Cloudflare); meta robots and X-Robots-Tag directives such as noindex or nosnippet.
Free check
Reading (35 points)
How much of the main text is in the HTML without JavaScript, compared with the page rendered in a real browser; status code, redirects and canonical; the sitemap and whether it is listed in robots.txt; response time.
Free check
Understanding (25 points)
Structured data (JSON-LD, Microdata, RDFa) validated against the schema.org vocabulary and the fields Google requires for Product, LocalBusiness and BreadcrumbList; structured data added only by JavaScript.
Free check (home page)
llms.txt (5 points)
Whether /llms.txt exists and follows the proposed format of llmstxt.org, and whether its links work.
Free check
Deep check
Up to 25 pages from your menu and sitemap, each rendered and checked: headings, language, hreflang, titles and descriptions, image alt texts, About / Contact / legal pages, authors and dates, FAQ sections.
Deep check (verified websites)
The check only reads public pages and files, like a visitor or a crawler does. It never logs in and never changes anything on your website.
How it works
How does the AI readiness check work?
You enter an address and pass a short automatic check against bots. Our server then reads your robots.txt, sitemap, home page and llms.txt with a clearly named scanner (EasySoftScanner/1.0), repeats the request with the User-Agent of each AI crawler, renders the page in a headless browser and shows the report on this page, usually in about a minute.
Enter the address. Please check websites you own or manage.
We read the rules. robots.txt is parsed exactly as RFC 9309 describes, for every official AI crawler, with the rule and line that applies.
We test the access. The same page is requested as a browser and as each crawler, so a firewall block or a challenge page becomes visible.
We compare and validate. The raw HTML is compared with the rendered page, and the structured data is checked against schema.org.
You get the report. A summary in plain words, what to do first, and the technical details with the evidence and the sources.
Reading the results
How do I read the results?
The crawlers are grouped by what they do, because blocking them has different consequences. Blocking training is a choice, not a problem: it does not lower your score. We do not tell you to allow everything; the report explains what each rule means so that you can decide.
Training Crawlers that collect content to train AI models, for example GPTBot, ClaudeBot or Google-Extended. Blocking them is legitimate and has no effect on search.
Search Crawlers that index pages for AI search and answers, for example OAI-SearchBot, Claude-SearchBot, PerplexityBot, Googlebot and Bingbot. Blocking them usually means your pages cannot be shown as a source.
On user request Fetches that happen when a person asks an assistant to open your page, for example ChatGPT-User or Claude-User. Some vendors state that these may not follow robots.txt.
Each result also has a confidence level. Confirmed means direct proof, for example the exact robots.txt rule. Probable means strong clues; the firewall test is never more than probable, because firewalls can also check the IP addresses of the real crawlers. Needs verification means only you can decide, for example whether a canonical link is intended. The score from 0 to 100 weighs access 35, reading 35, understanding 25 and llms.txt 5.
Example report
What does an AI readiness report look like?
Below is the real report of a test website from our lab. Its robots.txt blocks some search and user-request crawlers, a firewall rule blocks PerplexityBot, Cloudflare shows a challenge to another crawler, and there is no llms.txt and no structured data. It is not a customer website.
Example report Test site ai-block.corpus.test from our scanner test lab, built with deliberate robots.txt and firewall rules for AI crawlers. Not a customer website. Excerpt: 9 of 17 results (the informational ones are left out; the score is that of the full scan).
AI Readiness report · Basic scan
ai-block.corpus.test
Scanned on 30.09.2026 00:35 · http://ai-block.corpus.test/
81/100
How the score is calculated
Category
Points
Access for AI agents
21 / 35
Readable content
33 / 35
Structured data and identity
23 / 25
llms.txt
4 / 5
Scoring formula version ai-1. Each category starts with its maximum; points are deducted only for findings in this report.
Some things to improve
AI agents can reach the site, but some things make it harder to read or understand.
Score 81/100: the share of the checked points that are in order (higher is better). It covers what can be checked from outside, not how AI services decide to use your site.
Strong clues, but no direct proof. The reason is shown.
Needs verification
Cannot be seen from outside. Please check it as described.
Actively exploited
The flaw is on the CISA list of vulnerabilities that attackers are known to use. Deal with these first.
AI agents by purpose
AI training 5 of 8 allowed by robots.txt
Blocking AI training is a legitimate choice: it only says that your content should not be used to train AI models. AI search and answer engines use other agents.
Agent
Operator
robots.txt
Server access
GPTBot
OpenAI
BlockedDisallow: /line 5
OK
ClaudeBot
Anthropic
AllowedAllow: /$line 10
OK
Google-Extended
Google
BlockedDisallow: /line 5
Not tested
Applebot-Extended
Apple
Allowed
Not tested
CCBot
Common Crawl Foundation
BlockedDisallow: /line 5
OK
Meta-ExternalAgent
Meta
Allowed
OK
Amazonbot
Amazon
Allowed
Challenge (Cloudflare)
MistralAI-Training
Mistral AI
Allowed
OK
AI search and answer engines 10 of 11 allowed by robots.txt
AI search and answer engines may not be able to find, quote and link your site.
Agent
Operator
robots.txt
Server access
OAI-SearchBot
OpenAI
BlockedDisallow: /line 13
OK
Claude-SearchBot
Anthropic
Allowed
OK
Googlebot
Google
AllowedAllow: /line 28
OK
Google-CloudVertexBot
Google
AllowedAllow: /line 28
OK
PerplexityBot
Perplexity
Allowed
Refused (403)
Bingbot
Microsoft
Allowed
OK
Applebot
Apple
AllowedAllow: /line 28
OK
Meta-WebIndexer
Meta
Allowed
OK
Amzn-SearchBot
Amazon
Allowed
OK
DuckAssistBot
DuckDuckGo
Allowed
OK
MistralAI-Index
Mistral AI
Allowed
OK
User-triggered AI fetchers 6 of 6 allowed by robots.txt
When someone asks an AI assistant about your page, the assistant cannot read it for them. Some vendors say these fetchers may not follow robots.txt at all.
Agent
Operator
robots.txt
Server access
ChatGPT-User
OpenAI
Allowed
OK
Claude-User
Anthropic
Allowed
OK
Google-Agent
Google
No robots.txt name (identified by User-Agent only)
OK
Google-GeminiNotebook
Google
No robots.txt name (identified by User-Agent only)
OK
Perplexity-User
Perplexity
Allowed
OK
Meta-ExternalFetcher
Meta
Allowed
OK
Amzn-User
Amazon
Allowed
OK
MistralAI-User
Mistral AI
Allowed
OK
What we measured
Pages scanned
1
Text readable without JavaScript
/: 100%
robots.txt
found
llms.txt
not found
0Critical
0High
2Medium
5Low
2Info
Medium (2)
robots.txt blocks 1 AI search / answer engines agent(s) on the homepage
MediumConfirmed
What it means
Your robots.txt asks these agents for AI search and answer engines not to read your home page: OAI-SearchBot.
Why it matters
AI search and answer engines may not be able to find, quote and link your site.
What to do
If this is not intended, remove or narrow the rules for these agents in robots.txt (the exact rule is in the technical details).
Technical details
Confirmed: robots.txt fetched and evaluated with RFC 9309 group matching and precedence; the matching rule is quoted.
Blocked: OAI-SearchBot. Blocking search agents reduces the chance that AI search and answer engines (and, for Googlebot or Bingbot, classic search and the AI features built on it) can find, cite and link the site.
How to fix it
If you want to be found and cited by these services, remove or narrow the Disallow rules for these agents (exact rule per agent in the evidence). You can keep training agents blocked: the settings are independent.
Server/WAF refuses 1 AI search / answer engines agent(s)
MediumProbable
What it means
Your server or firewall refuses requests from agents for AI search and answer engines: PerplexityBot.
Why it matters
AI search and answer engines may not be able to find, quote and link your site.
What to do
If this is not intended, allow these agents in the firewall, CDN or security plugin, preferably by the IP ranges the vendors publish.
Technical details
Probable: Tested with the agent's User-Agent from the scanner's IP. WAFs and CDNs also verify real crawlers by IP address (reverse DNS or published ranges), so real crawlers may be treated differently: at most probable.
Requests carrying the User-Agent of PerplexityBot got an error (status per agent in the evidence) while a normal browser request got 200. Blocking search agents reduces the chance that AI search and answer engines (and, for Googlebot or Bingbot, classic search and the AI features built on it) can find, cite and link the site.
How to fix it
If this is not intended, allow these agents in your firewall/WAF/CDN bot rules (preferably verified by the vendors' published IP ranges, not only by User-Agent).
Some pages from your main menu are not in the sitemap.
Why it matters
Crawlers may find them later or not at all.
What to do
Add them to the sitemap, or check that they should indeed not be indexed.
Technical details
Confirmed: Compared the main-menu links with every URL in the sitemap.
Pages linked from the main menu but not listed in the sitemap: http://ai-block.corpus.test/about-internal, http://ai-block.corpus.test/news/?page=2, http://ai-block.corpus.test/private/press, http://ai-block.corpus.test/private/archive.
How to fix it
Add the pages to the sitemap (or check that they should not be indexed).
Optional. Major AI vendors have not confirmed that they use llms.txt, so the effect is uncertain.
What to do
Optional: publish /llms.txt with your site name, a short summary and links to your key pages.
Technical details
Confirmed: Fetched /llms.txt directly (the server answered with an HTML page).
http://ai-block.corpus.test/llms.txt was not found. llms.txt is a proposed standard (llmstxt.org); major AI vendors have not confirmed using it, and Google states it is not needed for Google Search. Impact uncertain.
How to fix it
Optional: publish /llms.txt with an H1 (site name), a short summary and lists of links to your key pages.
robots.txt blocks key pages for 7 AI search / answer engines agent(s)
LowConfirmed
What it means
Your home page is open, but robots.txt closes some important pages to agents for AI search and answer engines: /about-internal, /news/?page=2, /private/archive.
Why it matters
AI search and answer engines may not be able to find, quote and link your site.
What to do
Check that the rules for these pages are intended (the exact rule is in the technical details).
Technical details
Confirmed: robots.txt evaluated for the key pages linked from the homepage menu.
The homepage is allowed, but these agents are blocked on some key pages: Claude-SearchBot, PerplexityBot, Bingbot, Meta-WebIndexer, Amzn-SearchBot, DuckAssistBot, MistralAI-Index. Pages: /about-internal, /news/?page=2, /private/archive. Blocking search agents reduces the chance that AI search and answer engines (and, for Googlebot or Bingbot, classic search and the AI features built on it) can find, cite and link the site.
How to fix it
Check that the Disallow rules affecting these pages are intended (the exact rule is in the evidence).
robots.txt blocks key pages for 6 user-triggered AI fetchers agent(s)
LowConfirmed
What it means
Your home page is open, but robots.txt closes some important pages to agents for user-triggered AI fetchers: /about-internal, /private/archive.
Why it matters
When someone asks an AI assistant about your page, the assistant cannot read it for them. Some vendors say these fetchers may not follow robots.txt at all.
What to do
Check that the rules for these pages are intended (the exact rule is in the technical details).
Technical details
Confirmed: robots.txt evaluated for the key pages linked from the homepage menu.
The homepage is allowed, but these agents are blocked on some key pages: ChatGPT-User, Claude-User, Perplexity-User, Meta-ExternalFetcher, Amzn-User, MistralAI-User. Pages: /about-internal, /private/archive. These agents fetch a page when a user asks an AI assistant about it. Blocking them means the assistant cannot read the page for that user. Several vendors state these fetchers may not apply robots.txt at all (see evidence).
How to fix it
Check that the Disallow rules affecting these pages are intended (the exact rule is in the evidence).
Your pages contain no structured data (schema.org).
Why it matters
Structured data describes in a machine-readable way who you are and what the page contains.
What to do
Describe your organisation, your site and your main content types with schema.org (most SEO plugins can do it).
Technical details
Confirmed: JSON-LD, Microdata and RDFa extracted from the raw HTML.
Homepage: no JSON-LD, Microdata or RDFa with schema.org types was found.
How to fix it
Describe the organisation (Organization/LocalBusiness), the site (WebSite) and the main content types (Product, Article, BreadcrumbList) with schema.org JSON-LD.
Site served through Cloudflare: check the AI crawler settings
InfoConfirmed
What it means
Your site is served through Cloudflare.
Why it matters
Cloudflare has its own settings that can block or challenge AI crawlers, independently of robots.txt.
Technical details
Confirmed: Cloudflare response header observed.
Responses carry Cloudflare headers (cf-ray). Cloudflare can block or challenge AI crawlers with dashboard settings (Block AI Bots, AI Crawl Control), independently of robots.txt.
How to fix it
In the Cloudflare dashboard, review Security → Bots (Block AI Bots) and AI Crawl Control; decide per crawler (training vs search) instead of one global switch.
robots.txt: 21 of 25 AI agents allowed on the homepage
InfoConfirmed
What it means
21 of the 25 AI agents we know may read your home page according to robots.txt.
Why it matters
The table shows the decision and the exact rule for each agent, grouped by what the agent is used for.
Technical details
Confirmed: robots.txt fetched and evaluated per agent.
Each agent from the official vendor documentation was evaluated against robots.txt with RFC 9309 precedence (the exact matching rule per agent is in the evidence). Blocked: 4.
Where do the rules and the crawler list come from?
Only from official sources. The names, robots.txt tokens and purposes of the 27 crawlers come from each vendor’s own documentation; third-party bot lists are not used. A monthly job checks these pages again and reports every change to us for review, and each report shows when the list was last verified.
What does it cost? Free check, deep check with credits or monitoring
The basic check of the home page is free and needs no account. A deep check of up to 25 pages is available for websites whose ownership you have verified: the first one of each domain and account is free, after that it costs one credit. With Monitoring, the AI readiness check runs every month, together with the weekly security scan.
Free
Basic check
€0
Home page: robots.txt for every official AI crawler, firewall test, content without JavaScript, schema.org, sitemap, llms.txt
No account needed
Up to 3 scans per hour and 10 per day per website
Free deep scan One per module for each verified domain and each account (confirmed email address).
Pay per scan
Deep scan with credits
from €1.96 per scan
Pack
Price
1 credit
€4.90
3 credits
€9.90
10 credits
€24.90
25 credits
€49.00
1 credit = 1 deep scan of a verified website, on either module (security or AI readiness)
Deep AI check: up to 25 pages from the menu and the sitemap, headings, language, identity pages and FAQ
Credits are valid for 12 months from purchase
If a scan fails because of a problem on our side, the credit is returned
Subscription
Monitoring
from €7.90 per month
Plan
Websites
Manual rescans / month
Price
Monitoring Starter
1
—
€7.90 / month or €79.00 / year
Monitoring Agency
10
100
€24.90 / month or €249.00 / year
Monitoring Agency 25
25
250
€49.00 / month or €490.00 / year
Weekly security scan and an email when a newly published vulnerability affects a component we detected
Monthly AI readiness check
History and comparison of scans
Included rescans reset every month and are not carried over; cancel anytime, the plan runs until the end of the paid period
Credits and plans are bought in your GetEasySoft account. Customer accounts for the scanner open soon; the free scan is available now.
Prices in EUR.
Privacy
What happens to my data?
We store the address you checked, the results and the date, so that the report can be shown and shared. Your IP address is used only as a hashed value to limit abuse. The AI features that would send your content to an AI provider are switched off; the check itself uses no AI model.
Want it fixed without spending your evening on it?
We can adjust robots.txt and your firewall rules to the choices you make, add structured data, make the main content visible without JavaScript and write an llms.txt. Send us the address of your website, and you get a clear answer and a fixed price before we change anything.
Yes. The basic check of your home page is free and needs no account. There is a limit of 3 checks per hour per visitor and 10 per day per website. A deep check of up to 25 pages is available for websites you have verified: the first one is free, then it costs one credit.
Which AI crawlers are checked?
27 agents of 11 operators, taken only from the official documentation of each vendor: for example GPTBot, OAI-SearchBot and ChatGPT-User (OpenAI), ClaudeBot, Claude-SearchBot and Claude-User (Anthropic), Googlebot and Google-Extended, PerplexityBot, Bingbot (used by Copilot), Applebot, Meta, Amazon, DuckDuckGo and Mistral AI.
Should I let AI crawlers train on my content?
That is your decision, and blocking training crawlers is a legitimate choice: it does not lower your score. Blocking search crawlers usually means your pages can no longer be found and quoted as a source in AI answers. The report shows each purpose separately.
Does a good score mean ChatGPT or Google will recommend my website?
No. Nobody can promise that. The check measures technical conditions that you can verify and fix: whether AI crawlers may and can reach your pages, whether the content is readable without JavaScript and whether it is described with structured data.
Why does the firewall test say “Probable” and not “Confirmed”?
We send requests with the User-Agent of each AI crawler from our scanner and compare them with a normal browser request. Firewalls can also check the real IP addresses of the crawlers, which we cannot use, so a block or an allow is at most probable. Your server logs are the confirmation.
Is llms.txt required?
No. llms.txt is a proposed standard (llmstxt.org). The large AI vendors have not confirmed that they use it, and Google says it is not needed for Google Search. It has the smallest weight in the score: 5 of 100 points.
Do FAQ sections still help?
Clear questions with direct answers help people and AI assistants understand a page. The FAQ rich result in Google is a different thing: Google limited it in 2023 and no longer shows it since 7 May 2026. The report says both.
My website uses Cloudflare. What should I check?
Cloudflare can block AI crawlers with “Block AI Bots” or AI Crawl Control, and it can show a challenge page instead of your content. When the check detects Cloudflare or a challenge, the report links to the official Cloudflare settings.
Does the check change anything on my website?
No. It only reads public pages and files, like a visitor or a crawler does: robots.txt, the sitemap, your pages and llms.txt. It never logs in and never sends forms.
How much does a deep check cost?
The first deep check of each verified domain and account is free (confirmed email address). After that one deep check costs one credit, from 4.90 EUR for one credit to 49 EUR for 25. Credits are valid for 12 months and also work for the security scanner.
Check your website now
The free check takes about a minute. If something needs changing and you do not have the time, we can do it for you.
