If an audit gets 403, first restore its access to the relevant public page. As long as the site’s HTML is replaced by a block screen, warnings like “missing title” or “missing H1” do not describe your page. The fact that the site opens in the owner’s browser does not explain the denial either: the conditions for the two requests may differ.
Start with the exact URL, the time of the check, and the block message. This information will help your hosting provider find the rule that was triggered.
Open the page without signing in
Open a new private window without your session and visit the same URL. The administrator may be allowed in while a visitor is asked to complete a browser check. Note whether such a check appears, where the site redirects you, and what response the document receives in the Network tab.
Ask a developer to save a separate HTTP request. In a terminal, the command looks like this:
curl -sS -D headers.txt -o response.html 'https://example.com/catalog/'In headers.txt, look for the status and request ID; in response.html, look for the denial message. Replace the domain with yours. This is another client, not an exact reproduction of a reChecker request: its connection address is different, and its headers may differ.
Find the denial in the site’s security logs
Security systems may filter connection addresses, request frequency, country, headers, or clients that do not execute JavaScript. A WAF is a filter for requests in front of an application. The issue to investigate is the specific trigger, not the general assumption that “the site blocks bots.”
Compare one ordinary request with the audit result. If the first request succeeds but crawling several pages stops, a rate limit may be involved. If the first request is denied, the cause may be different. Do not deliberately generate a large volume of requests: it is enough to share the event you already have with your security administrator.
You can write this in your support message:
An audit of the public page https://example.com/catalog/ received 403 4 October at 12:10 MSK. It opens in a browser without signing in. The response includes request ID abc-123. Please find the blocking rule that was triggered at that time and let me know how to allow checks of this public section.
This is a filled-in example: replace the address, time, and ID with your own. Cookies, tokens, and a full browser log are not needed for the initial support request.
Adjust the rule that is blocking the check
If a rate limit is confirmed, agree on a slower crawl or the necessary exception. If the connection address is blocked, use the official method provided by your security system to allow the client. The scope of the change should cover only the relevant public pages.
You do not need to disable all site protection for a single audit. However, a simple User-Agent string can be supplied by any client and does not prove its identity. The developer or hosting provider should explain exactly what is allowed and what criterion is used.
If the site requires a browser check that the tool cannot pass, ask both parties what verification method is supported. There is no need to keep retrying endlessly to get around a denial.
Check Google’s access separately
403 from a third-party auditor does not prove that Google receives the same response. Check the URL inspection details in Search Console and the relevant server logs. You can verify genuine Google requests using the official instructions; spoofing the Googlebot name is not a substitute.
Rerun the audit and open the page it retrieved
After the rule is changed, the tool should receive the HTML for the relevant section. A 200 response that says “confirm that you are human” is still not a catalog page. Check the content, then review the results for metadata and links.
Also open the private account area: it should still require a login. Save the outcome of your support request and the date of the successful audit. Further checks are available in reChecker’s technical audit; the status code reference and SEO audit process can help you interpret the results.