20 questions. 16 right earns the certificate.

CrawlCheck Certified · the test

20 questions. 16 right earns the certificate.

Every question is answered in the six lessons. Your answers are sent only when you press Submit; a pass stores your name, the date and the course version against a verification code, and emails you the certificate link. A reload draws a different set.

Name as it should appear on the certificate

Email (the certificate link is sent here; stored only as a hash)

1. Which agent fetches one page live because someone asked about it? CCBot
PerplexityBot
Perplexity-User
Bingbot

2. A well-behaved crawler reads a Disallow for itself in robots.txt. What happens? The server returns 403
The crawler is shown a challenge page
The crawler does not fetch the path
The crawler fetches it but does not index it

3. Why not re-scan from a logged-in browser? It is slower
It bypasses most edge rules, so it does not see what a crawler sees
It costs credits
Browsers cannot read robots.txt

4. A file has a * group with Disallow: /private/ and a later GPTBot group with Allow: /blog/. May GPTBot read /private/? No: the * rules apply to everyone
Yes: it obeys only its own group
Only if the groups are separated by a blank line
Only after the sitemap is read

5. Which group does a crawler obey? Every group in the file, merged
The first group in the file
The most specific group that names it
The * group, always

6. Which crawler builds the index an assistant answers from? ClaudeBot
Claude-SearchBot
Claude-User
Applebot-Extended

7. Where should the business name, phone and address agree? Only on the contact page
On the visible page, in the structured data and in the directories the site links to
Only in the schema
Only in Google Business Profile

8. What is the signed answer? The page's HTML
The record an AI agent gets about a site before it reads, cites or buys
A PDF of the report
The sitemap

9. Why check both example.com and www.example.com? Search engines only read www
A redirect on one and a different file on the other is a split nobody notices
robots.txt is only read on the bare host
Because llms.txt lives on www

10. A robots.txt has only a * group with Disallow: /admin/. Does ClaudeBot obey it? No: ClaudeBot is not named
Yes: with no group naming it, ClaudeBot obeys *
Only if a Content-Signal is present
Only on the www host

11. How many values does a Content-Signal line carry? One
Two
Three: search, ai-input, ai-train
Five

12. Which of these is NOT a place a crawler can be stopped? robots.txt
A CDN bot rule
A firewall
The page's title tag

13. Which of these is NOT in the signed answer? Who controls the domain
What the site declares
What it actually served
The owner's email address

14. Which is the most common machine-file failure? A file that is too large
A login, challenge or not-found page served with status 200
A missing favicon
A sitemap with more than 100 URLs

15. Two phone numbers appear on one site. How does a retrieval system read that? As one business with two lines
As two businesses
As a typo it corrects
It ignores phone numbers

16. Which tool runs the business record check? crawlcheck.io/listing-check
robots.txt
The sitemap
llms.txt

17. What should you fix first when the record check shows disagreements? The directory listings
Your own site
The Google listing
The competitors' listings

18. A site's /llms.txt answers 200 with the theme's 'not found' page. What is wrong? Nothing: 200 means the file exists
The status says the file exists and the body says it does not
llms.txt should answer 404
The sitemap is missing

19. Where must a Content-Signal line be written to bind every agent? Once at the bottom of the file
In every group
Once at the top of the file
In the sitemap

20. What does search=no mean? The content may not be found and shown
The content may not train a model
The sitemap is disabled
Only Google may index it