About the Securick scanner
Who is SecurickBot?
SecurickBot is the scanner behind Securick. When someone scans a site on securick.com, it loads that site's public pages the way a visitor would and reports security and launch problems to the person who asked. Every request uses this user agent:
SecurickBot/0.1 (+https://securick.com/scanner)
Deep scans run only on sites whose owner has verified the domain with us. Their requests also carry the header x-securick-scan: deep, and the user agent ends with (Securick deep scan).
What it requests
- Only ports 80 and 443 (HTTP and HTTPS), and only public addresses: never private or internal networks.
- A free or Pro scan only reads: the home page, its own scripts and source maps, and a list of well-known paths (such as
/.envor/.git/config) to check that private files aren't public. It sends no logins and changes nothing. - Deep scans, on verified domains only, also send a small number of harmless test requests to the site and its database API, with the owner's consent.
- Every scan stops after 5 minutes (10 for a deep scan).
Two requests that look like Google
To catch hacked sites that show spam only to search engines ("cloaking"), our malware and spam check loads the home page twice more per scan: once with Google's crawler user agent (Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)), and once as a visitor arriving from Google search (Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/130.0.0.0 Safari/537.36 with the referrer https://www.google.com/). We compare what those see with what SecurickBot saw. These two loads are never used to get past a firewall: if your firewall blocked SecurickBot, the scan has already stopped.
What we never do
- We never disguise the scanner to get past a block, solve a captcha or retry with other headers. If your firewall turns us away, the report says so in plain words and the scan doesn't count against anyone's free scans or Pro pass.
- On a site nobody has verified, we never send sign-in attempts or bursts of requests, and we never write to a site unless its owner verified it and switched that test on.
How to let Securick through your firewall
Best: your own allow header. A user agent can be copied by anyone, so a rule that allows "SecurickBot" lets strangers through too. Instead, verify your domain with us: on your report, open "Deep scan", then "Let Securick through your firewall", and press "Show my allow header". You get a secret value for your site only. Scans you start send it as x-securick-allow: <your value>, only to your site, only over a working HTTPS connection, and never on scans other people start. Allow that header in your firewall:
- Cloudflare: Security > WAF > Custom rules: a rule with the expression
(any(http.request.headers["x-securick-allow"][*] eq "<your value>"))and the action Skip (skip the bot fight mode, the managed rules and rate limiting). - AWS WAF: a rule with a higher priority than your bot rules that matches the header
x-securick-allowexactly equal to your value, with the action Allow. - Akamai, Imperva, Azure Front Door, F5, Sucuri and others: an exception (allow or bypass) for requests whose
x-securick-allowheader equals your value, or ask your provider's support. - WordPress (Wordfence and similar plugins): these usually can't match a header: allow the user agent
SecurickBotwhile you scan, then remove it.
Keep the value private. If it leaks, press "Make a new value" on the same panel: the old one stops working at once. It also stops being sent when your domain verification ends (after 30 days); verify again from the same account and the same value comes back.
Quick but weaker: allow requests whose user agent contains SecurickBot, only while you scan. Anyone can copy a user agent, so remove the rule afterwards.
Then scan again. Seeing SecurickBot on a site you didn't scan, or want it to stop? Tell us through the form on our contact page and we'll look into it.