Documentation
Everything you need from sign-up to troubleshooting. If a topic is missing, open a ticket and it gets added.
Getting started
HexShield is a hosted service: there is nothing to install on your server. Your traffic is proxied and filtered on the HexShield edge, and clean requests are forwarded to your origin.
- Sign up. Create an account with the Get started button. The account is free; protecting a domain requires Pro ($50/month, up to 2 domains), which you buy inside the panel in crypto via NOWPayments.
- Add your domain. In the panel choose Add Site and enter your domain together with the public IP of the server that currently hosts it (your origin).
- Point your A record. At your DNS provider, set the A record of your domain (and of www) to the HexShield edge IP shown in your panel.
- Adjust the origin (optional). If your origin moves or listens on another port, update it later from the panel — no DNS change is needed for that.
- Done. Once DNS propagates, the certificate is issued automatically and traffic to your domain is filtered before it reaches your origin.
DNS and SSL
Add your domain and origin IP in the panel, then point your domain's A record at the HexShield edge IP shown in your panel. Once DNS propagates the certificate is issued automatically through Let's Encrypt and renewed automatically.
From that point your origin IP is no longer visible from outside. Remember to clean up old A records, stray subdomains and MX/SPF entries pointing at the origin — that is the most common way an origin IP leaks.
# Your DNS records
example.com. A <HexShield edge IP from your panel>
www.example.com. A <HexShield edge IP from your panel>
# Do not leak:
# direct.example.com A <origin IP> <-- DELETE THIS
# mail.example.com A <origin IP> <-- move to a separate host
Using the panel
The panel has a sidebar with Overview, Analytics, Security (visitor verification, custom verification screen, Managed Protections, WAF and custom rules, access control, origin IP hiding, Proof-of-Work difficulty, attack alerts, threat radar), DNS (records, analytics, settings), SSL/TLS, Speed and caching (cache, optimization, redirect rules), Maintenance, Account and Support tickets. Every WAF switch carries a description of what it does and what it risks.
Every change reaches the edge instantly — no restart, no waiting. After a save the edge cache for that domain is purged immediately.
Analytics has five tabs — Traffic, Security, Performance, Geography and Top lists — with time ranges from 30 minutes to 30 days. The Overview page and the Geography tab show a world map of where requests come from. In Geography you can switch between Requests and Attacks: Attacks counts only requests the edge blocked as attacks (exploit signatures, scanner tools, floods and rate limits, banned addresses), by source country. Verification screens, your own country / IP / Tor / VPN / custom rules and traffic dropped in the kernel after a ban are not included, and very high rates are sampled.
Verification settings
There are two separate settings and they are easy to confuse: “Verification” decides WHEN the screen appears (Off / Only during an attack / Always), “Screen” decides HOW it looks (automatic / checkbox). To turn verification off entirely, set the first one to Off; the second has no effect then.
Both screens use JavaScript in the background — that is what proves the visitor is a real browser. Choosing the checkbox does not weaken protection; what the server verifies is the same proof-of-work in both modes.
Country and IP rules
Two modes: “block listed countries” and “allow ONLY the listed ones”. The second is powerful; saving it with an empty list is refused, because that would block everyone — including you.
Allow-listed IPs are exempt from every check: rate limits, WAF scanning and verification are all skipped. Adding your own office address and your monitoring services is recommended.
By default a blocked visitor only gets the 403 page. A platform administrator can switch on “Ban blocked visitors at the kernel” for a site: the first request still gets the 403 page, then the visitor’s address is banned for 15 minutes and the Linux kernel drops its web traffic before the web server sees it (visitors behind a CDN are banned at the edge only). The ban is server-wide — the address is dropped on every site hosted here — and administrators, trusted addresses and search-engine bots are never banned. While it is on, only the administrator can widen or change the site’s country rules; the site owner can only loosen them (remove a blocked country or, in allow-only mode, add an allowed one).
Cache levels
Four levels: off, easy (static assets only), medium (static plus a 60-second HTML micro-cache for visitors carrying no cookie of their own) and hard (static plus the micro-cache for anyone without a session cookie).
Never cached at any level: POST and other non-GET requests, anything carrying an Authorization header, and personal paths such as /admin, /account and /api/ — even when the URL ends in .png or .js.
Speed and Rocket Loader
The Speed and caching section has three pages: Cache (cache level, purge tools and development mode), Optimization and Redirect rules.
Rocket Loader, on the Optimization page, is off by default and is switched on per site. When it is on, HexShield changes blocking <script> tags to the browser’s own “defer” mode, in their original order, so the browser can draw the page while the scripts download. Pages it cannot change safely keep the HTML exactly as your server sent it: pages with a Content-Security-Policy, scripts that use document.write, payment, captcha and cookie-consent scripts, and pages where the order cannot be kept.
The “Never defer these scripts” box takes one word per line (up to 50): a script whose address, id or code contains the word is left alone. Automatic recovery is on by default: if an optimized page raises a script error, that visitor’s page reloads once without Rocket Loader. To compare, add ?hx_speed=off to a page address to load that page without Rocket Loader.
Rocket Loader does not make your own server faster: the first visit to a page that is not in the edge cache still waits for your server (see Cache levels). While it is on, the HTML pages it looks at are held at the edge until they are complete so they can be checked, and are sent without ETag and Last-Modified headers. After switching it on, check your important pages in a private window, and turn it off again if anything misbehaves.
Troubleshooting
If your site is not behaving as expected, first check that your domain's A record points to the HexShield edge IP shown in your panel, then open the Performance tab of Analytics in the panel to confirm HexShield can reach your origin. If both look right, open a ticket in the panel.
If you suspect something about the verification page, check the X-HexShield-Challenge response header in your browser's network tab: it says captcha, js or js_fallback. If you see js_fallback, open a ticket in the panel and mention that header.
# Does your domain point at the HexShield edge?
dig +short example.com A
# -> should print the edge IP shown in your panel
dig +short www.example.com A
Security notes
Use a strong, unique password for your HexShield account. If you suspect someone else has access, change it and sign out all sessions from your profile in the panel. Add your own office address and your monitoring services to the IP allow list so they are not challenged.
Once your site is behind HexShield, restrict your origin's firewall so ports 80 and 443 only accept connections from the HexShield edge IP. Otherwise an attacker who learns your origin address can go around the protection.
Account security and notifications
If the operator has turned email delivery on: use “Forgot your password?” on the sign-in page to get a reset link (valid for 30 minutes; it never turns off two-factor authentication, so you still need your code). New accounts confirm their email address first; until then you can look around, but you cannot add domains, pay, open support tickets, create API tokens or set an email alert channel.
Account > Security shows your email address and lets you change it: you enter your password (and your two-factor code), and the new address has to confirm the change with a link. The old address gets a notice with a link to cancel the change. Changing the email address, or resetting the password, signs out every device.
Account > Activity log lists the security-relevant actions on your account (sign-ins, password and email changes, site, origin, rule and cache changes, API tokens) for 180 days (latest 5,000 entries). Filter by category, domain, text or dates and download the result as CSV.
Account > Notifications chooses the language of account emails and turns the subscription-expiry, payment and support-reply emails on or off; security notices are always sent. Site events (origin down or back up, certificate expiring or renewal failed) are off by default: switch them on per site under Security > Attack alerts. “Send sample” first sends an example message to your channels.
Account > Security > “Confirm sensitive changes” is off by default. When you turn it on, changing the origin address, adding an IP to the allow list or changing an alert channel asks for your password (and two-factor code) first, and stays unlocked for a few minutes afterwards. An alert email address other than your account address becomes active only after its owner opens the confirmation link we send to it.
API
Create a token under Account > API tokens. Send it as Authorization: Bearer <token> to https://<panel host>/api/v1/. Tokens carry limited permissions (sites:read, analytics:read) and can be limited to chosen sites.
Limits: 120 requests per minute per token. Errors are JSON with a code and a message. The origin IP is never returned. The machine-readable description is at /api/v1/openapi.json.
Keep tokens secret. Revoke a token in the panel if it leaks; resetting your password by email revokes all tokens.
curl -H "Authorization: Bearer hxs_xxxxxxxx_..." -H "Accept: application/json" https://panel.example.com/api/v1/sites
Put your site behind the shield today
Nothing to install. Point your DNS at the HexShield edge, keep your origin IP out of sight, and filter attacks before they reach your server.