Scan 5 pages, free
Automated testing of five pages, everything it finds, and an honest account of what it cannot find. No account. No card.
What arrives in the email
- The 5 URLs that were scanned
- Every issue found on them, not a top three
- The WCAG success criterion, the rule id and an example element
- What was not checked, and why
- The engine and version, and the date and time it ran
What it cannot find
Automated testing detects roughly 30–40% of real barriers. Manual testing covers the rest.
16 of the 50 WCAG 2.1 AA success criteria have any automated coverage — 32%. Roughly two out of three real barriers require a human.
The limits, in numbers
5 pages · 1 scan per website per day · 3 per network per day · nothing kept after 48 hours · robots.txt obeyed in full
- Pages per scan
- 5
- Scans per website
- 1 per 24 hours
- Scans per network
- 3 per 24 hours
- Scans in total
- A hard daily cap across the whole service
- Result retained
- 48 hours, then deleted
- Verification
- None. This is the one scan that runs on a domain nobody has verified, and that is why it is five pages and not five thousand
- Politeness
- robots.txt obeyed in full including Crawl-delay; one request at a time; 2 seconds between requests; aborts after three consecutive 5xx responses
- User agent
- BarrierOpsBot/1.0 (+https://barrierops.com/bot)
The 2028 checklist, with no email address
The Government Website Accessibility Checklist 2028, as a tagged PDF and as a web page. No form, no address, no “we’ll send you a link”.
How the five pages are chosen
- The address you gave, exactly as you gave it, after following redirects within the same registered domain — at most two.
- Then up to four more, chosen from the links in that page’s main navigation, in document order, taking the first four that are on the same registered domain, return HTML and are allowed by robots.txt.
- If fewer than four qualify, fewer than five pages are scanned, and the result says how many and why.
- Nothing is discovered by crawling. There is no second level. The list is fixed before the first request goes out.
Five pages chosen this way are usually four templates and a home page. That is enough to find issues that repeat across a site and not nearly enough to describe one.
What we do to your server
- Five GET requests, one at a time, two seconds apart, from BarrierOpsBot/1.0.
- robots.txt is read first and obeyed in full, including Crawl-delay.
- No form is submitted, no button is clicked, nothing is written. It only issues GET requests.
- To stop it — now and permanently — use the form on our crawler page.
Questions we are asked
Does a clean result mean our site is compliant?
No. Automated testing evaluates a minority of the WCAG 2.1 AA success criteria, and this scan looked at five pages. A clean automated result is a good sign about those five pages and nothing more.
Why only five pages?
Because nobody has verified that you own this website. Five pages is the most we are willing to fetch from a domain on an anonymous request. A verified property is scanned in full.
What happens to my email address?
It receives the result and is used to answer you if you reply. It is not added to a list, not sold, not shared, and there is no list to unsubscribe from. Ask us to delete it and we will.
Can we scan the same site again tomorrow?
Yes. One scan per website per day, and three per network per day.
Can you check our PDFs?
Not in the free scan. Documents are validated against PDF/UA-1 with veraPDF in the product, per document, naming the clause that failed.
What does the paid product do that this does not?
Clusters thousands of URLs into templates, evaluates every template on a schedule, names the URL regions it sampled rather than exhausted, records manual testing beside automated results, and keeps an append-only evidence ledger you can export and verify. Pricing is published in full.