AdGuard DNS typosquatting protection filtering suspicious lookalike domain requests

AdGuard DNS Typosquatting Protection: 7 Smart Checks

AdGuard DNS typosquatting protection can block suspicious lookalike domains before your browser connects, but it cannot identify every fake website. The feature checks for deceptive spellings such as missing, added or swapped letters. A separate AdGuard DNS setting addresses some internationalized domain name tricks known as homograph attacks. Neither replaces a browser’s security warnings, careful account verification or phishing-resistant sign-in.

This guide explains how to activate both security controls in Private AdGuard DNS, confirm your device is using the correct resolver, investigate blocked requests, and repair false positives without weakening everything else. You will also learn when DNS phishing protection fails because an attacker uses an otherwise legitimate domain.

I have organized the workflow into seven checks. A security setting that stops a suspicious domain is useful. A setting that has never been connected to the actual device is mostly a decorative checkbox.

AttackTypical trickDNS limitation
TyposquattingMissing, extra or swapped lettersLegitimate names may look similar
IDN homographLookalike letters from different scriptsNot every deceptive domain is detected
Phishing on a trusted hostMalicious content at a legitimate domainDNS cannot inspect page paths

Key Takeaways

  • AdGuard DNS version 2.21 introduced typosquatting protection, which analyzes suspicious variations of domain names.
  • Version 2.22 introduced a separate control for homograph attacks, disabled by default according to its release documentation.
  • The controls described here are in the Private AdGuard DNS server dashboard; a public DNS address does not provide the same management interface.
  • Browser Secure DNS, IPv6, operating-system settings and VPN resolvers may bypass an expected DNS policy.
  • A blocked domain deserves investigation. Narrow allow rules are safer than disabling an entire protection layer.
  • DNS phishing protection is preventive filtering, not proof that a website or a payment request is legitimate.

Affiliate disclosure: HackersGhost may earn a commission through sponsored links without increasing your purchase price. The Private AdGuard DNS offer is optional. You can understand every security recommendation in this article without buying anything.

Eligible one-year plans only. Code HACKERSGHOST20 applies automatically; check the final price before payment. Public promotions may offer similar or different pricing.

How fake domains fool people before a website loads

Phishing email illustrating a deceptive website identity and a fishing hook

A convincing phishing page can look almost identical to a trusted login portal. The attacker may reproduce colors, logos and a familiar layout, but the hostname reveals where the browser actually connects. An extra letter in a domain, a missing hyphen or an unusual character can change the owner of that destination. The page may still use HTTPS because criminals can obtain certificates for domains they legitimately register.

Typosquatting protection is designed for a particular part of this problem. It examines spelling-based similarities in domain names rather than trying to judge the artwork on a phishing page. AdGuard says its classifier looks for patterns such as swapped characters, omissions and insertions. A domain that resembles a popular site is not automatically malicious, which makes false positives an unavoidable consideration.

Homograph attacks are related but technically different. An internationalized domain can contain a character from another alphabet that visually resembles a Latin letter. Internationalized domain names are legitimate and important; the abuse lies in deliberately imitating a trusted destination. The Internet Corporation for Assigned Names and Numbers (ICANN) publishes broader information on internationalized names and DNS abuse.

Why a familiar-looking address is not enough

A browser may display some internationalized domains using an ASCII-compatible form starting with xn--. That prefix is not a malware verdict. Many real, legitimate multilingual domains use internationalized names. Conversely, a completely ordinary-looking ASCII domain can still be malicious. Check the complete hostname and the context of the request, rather than trusting one visual clue.

DNS filtering operates at the hostname level. It can block a domain associated with phishing; it generally cannot distinguish a malicious login page from an innocent page when both live under the same allowed hostname. This limitation is why typosquatting protection is one useful boundary rather than a complete anti-phishing system.

1. Confirm which kind of AdGuard DNS you use

Before changing a switch, distinguish the free public DNS resolver from Private AdGuard DNS. Public DNS can provide filtering with minimal setup. Private DNS offers a management dashboard with servers, device associations, security settings, Query logs and User rules. The following controls are documented for the private service, and account tiers may affect available management capacity.

Do not confuse this with AdGuard Home, the self-hosted resolver software, or with the separate AdGuard Ad Blocker application. They solve overlapping problems but do not necessarily expose the same proprietary hostname similarity features. A router that points to public AdGuard DNS addresses does not automatically get your Private DNS account configuration.

If the screen in front of you has a list of managed servers, open the server serving the device you plan to protect. If not, establish which AdGuard product you are running before trying to follow instructions intended for another platform. The first step in typosquatting protection is knowing where the filtering decision happens.

Make an initial inventory

  • Write down the intended device and network: laptop, phone, router or a small test group.
  • Identify whether it uses public DNS, a Private DNS endpoint, AdGuard Home or an application-level DNS feature.
  • Note whether a VPN or browser-specific encrypted DNS is active.
  • Check which server or profile owns the policy and which administrators may change it.
  • Keep a simple rollback note before modifying any security settings.

A misleading success signal is seeing advertisements disappear while assuming the same policy handles suspicious domains. DNS filtering may be active on one resolver while your intended typosquatting protection setting belongs to another. Verify the actual path, not just the visual result.

2. Enable typosquatting protection in the private dashboard

AdGuard introduced automatic typosquatting protection in DNS v2.21. In the private dashboard, open Servers, select the relevant server, enter Security, and enable Typosquatting protection. These are the documented navigation labels; the interface may change, so follow the server-specific Security area if your screen differs.

This control is intended to block some suspicious lookalike domains before a connection is established. It can improve protection against misleading links in phishing emails, fake login screens and deceptive advertisements. It does not know whether the person requesting a payment really works for the organization they claim to represent. It also cannot determine that a correctly spelled site has not been compromised.

Record the server name, switch state, device and time after applying the change. If your account has more than one private server, check the others separately. Typosquatting protection on your main laptop server should not be assumed to govern a phone using another configuration.

Heuristic detection versus a phishing blocklist

A conventional security blocklist tries to stop domains already identified as unwanted. Similarity-based typosquatting protection can recognize certain spelling patterns, including variants that were not explicitly listed in a traditional blocklist. Neither system is omniscient. Attackers constantly register new domains, and legitimate businesses can have names that resemble better-known sites.

The sensible configuration is layered: use appropriate security lists, keep browser protection enabled and add similarity checks where available. Do not remove a known-good security filter because you have found a new button with a more exciting label. Security theater has enough props already.

Magnifying glass identifying deceptive lookalike domain symbols in homograph attacks

3. Turn on the separate defense against homograph attacks

AdGuard DNS v2.22 added a feature that detects certain IDN homograph attacks. It is explicitly documented as disabled by default. On the chosen Private DNS server, go to Security → IDN homograph attacks protection and enable the setting if it fits your environment. Do not assume that switching on typosquatting protection automatically enables this second defense.

The distinction matters. A classic typo might add or omit a letter; a homograph can look correct while containing a different Unicode character. A malicious sender might exploit this similarity in a message or a forged login page. Because multilingual domains are common and legitimate, the goal is to catch suspicious impersonation rather than reject every name written in another language.

When discussing homograph attacks with colleagues or readers, use non-clickable explanatory examples rather than real malicious domains. You do not need to register a deceptive brand imitation or enter credentials into a suspicious site to understand the threat. If a browser shows an xn-- form, treat it as a prompt to inspect the hostname, not as evidence of a confirmed attack.

What these controls cannot promise

Typosquatting protection and homograph defenses are classification tools. They cannot guarantee a block for every confusing name, every top-level domain or every newly registered phishing site. If a particular domain is not blocked, confirm the device used the correct resolver before concluding that a feature failed. Even with the correct path, a detection miss remains possible.

Router sending network requests through a protected DNS resolver

4. Verify DNS phishing protection reaches your device

A server policy helps only when the device sends its DNS queries to that server. On a desktop or phone, DNS may come from router DHCP, a manual operating-system setting, a browser’s Secure DNS feature, a VPN tunnel, or a mobile private DNS configuration. IPv4 and IPv6 may also use different paths. DNS phishing protection becomes a routing question before it becomes a detection question.

  • Open the Query log for the intended Private DNS server and note the current time.
  • From the test device, make an ordinary request for a trustworthy hostname you use regularly.
  • Look for a corresponding event in the log, accounting for cached DNS results.
  • If nothing appears, check the browser’s Secure DNS choice, VPN-provided DNS, operating-system configuration and router DHCP settings.
  • Check IPv6 as well as IPv4. Confirm the expected policy applies to this specific device, not merely to another machine on the network.

A missing event is not necessarily a failed resolver: your device or browser may have cached the answer. Repeat with a harmless, authorized lookup and compare timestamps. A single router log entry also does not prove every application on every connected phone uses the router’s resolver. Avoid sweeping conclusions from one convenient screenshot.

Encrypted DNS and VPNs solve different problems

DNS-over-HTTPS, DNS-over-TLS and DNS-over-QUIC encrypt queries between supported clients and resolvers. They do not prove the website itself is trustworthy. A VPN encrypts traffic through a tunnel and may supply its own DNS. If that resolver bypasses your Private DNS server, the expected anti-phishing settings might not apply. Do not disable a VPN’s leak protection merely to make your Query log appear busy.

A certificate and HTTPS padlock mean your browser established an encrypted connection to the displayed hostname. They do not independently verify that the hostname belongs to your bank. Typosquatting protection is valuable precisely because attackers can register plausible-looking names and serve polished HTTPS websites.

HackersGhost Note: In my lab, I separate the configured resolver from the one a device actually uses. I prefer a repeatable DNS-path check to a hopeful green toggle. More RAM, more extensions and a louder browser warning cannot correct a query sent to the wrong service.

For a dedicated walkthrough of resolver checks, use the established HackersGhost DNS guide below. This article deliberately focuses on deceptive hostnames rather than repeating the broader privacy and router setup process.

AdGuard DNS Test: 7 Powerful Checks for Better Privacy

Verify your actual resolver and filtering path here before judging whether a new anti-phishing feature works.

5. Test typosquatting protection safely

A safe test does not involve visiting random lookalike login pages. Confirm the settings, inspect Query log traffic from ordinary browsing, and investigate a blocked request when one arises. If a domain is blocked, the log may reveal a reason, but not every interface or version necessarily exposes a precise classifier label. When that detail is missing, record the cause as unverified.

You can test the broader filtering pipeline by creating a temporary blocking rule for a harmless domain that you own or are authorized to administer, checking the blocked result, and removing the rule. That proves your DNS path and custom filtering behavior. It does not prove the proprietary typosquatting detector recognized a fake site. Keep those results separate in notes and screenshots.

If the dashboard reports a similarity-related block during your normal browsing, capture only the relevant details: the server or profile, time, hostname and available block reason. Avoid publishing full raw Query logs, since they may reveal browsing patterns, internal hostnames and device identifiers. Useful evidence should not create a new privacy incident.

Interpret a block accurately

A DNS request can be blocked by a security list, a custom user rule, parental controls, typosquatting protection or another policy. Seeing a red status does not identify which one fired. A successful manual block is a positive control for filtering; a confirmed classifier hit is evidence about that classifier. Conflating them makes a clean report impossible.

A missed detection also has several explanations: DNS bypass, a cached answer, the wrong private server or an unrecognized domain variation. Do not claim a universal hit rate, and do not invent detection thresholds. The product documentation confirms the feature exists, not that it will identify every suspicious hostname you can imagine.

6. Investigate false positives without disabling security

Similarity detection sometimes flags legitimate sites whose domains resemble a better-known name. AdGuard explicitly warns about false positives. A blocked invoice system or customer portal may be inconvenient, but it is not a reason to remove all typosquatting protection from every device. Start with independent identity verification, not with the link that was blocked.

  • Check the expected hostname through a trusted bookmark, official vendor documentation or another independent contact channel.
  • Find the exact request and time in the Private DNS Query log.
  • Determine which policy blocked it, if the interface provides that detail.
  • For a confirmed legitimate hostname, use Servers → Server settings → User rules → Add new rule to create a narrowly scoped allow rule using the supported rule syntax.
  • Retest only the affected service, record why the exception exists, and review it periodically.

Avoid broad wildcard allow rules that permit every subdomain of a provider. They may disable blocking in places you did not intend. If a service still fails after an exception, check browser and OS caches, related hostnames and other active policies. Change one thing at a time, or the next incident will become an archaeological expedition through forgotten settings.

When a legitimate internationalized domain is blocked

First establish whether an IDN homograph setting, a conventional security filter or a user rule caused the block. Legitimate internationalized domains are part of the Internet, not an attack category. If evidence remains ambiguous, collect sanitized details and consult the service provider instead of allowing an entire class of domains by default.

7. Add protections DNS cannot provide

DNS phishing protection is helpful for deceptive hostnames, but it cannot inspect page text, distinguish safe and malicious URL paths on the same allowed host, or validate a caller’s authority. Attackers can use compromised legitimate websites, file-sharing services and convincing emails. Your account security must assume that some malicious content will pass DNS filtering.

The Cybersecurity and Infrastructure Security Agency (CISA) offers practical public resources on phishing defense. The underlying habits still matter: verify unexpected messages through an independent channel, avoid entering credentials after following unsolicited links, and use stronger account authentication.

  • Use passkeys or phishing-resistant security keys for important accounts where available.
  • Use a password manager to reduce accidental credential entry on unfamiliar login origins.
  • Update browsers, operating systems and endpoint protection.
  • Confirm financial or credential requests through an independently obtained contact route.
  • For business domains, consider domain monitoring and email authentication controls as separate defenses.
  • Teach colleagues that a genuine HTTPS connection is not proof that a payment demand is genuine.

There is another boundary: typosquatting protection on your devices cannot prevent an attacker from registering a confusing domain elsewhere. Domain-registration monitoring and brand-abuse response address that problem. Preventing resolution and preventing registration are different security objectives.

What about self-hosted AdGuard Home?

AdGuard Home is self-hosted DNS filtering software with its own update, rule and infrastructure responsibilities. Do not assume its feature list matches the managed Private DNS interface just because both products share the AdGuard name. Choose between them based on the real administration model you need, and verify any required proprietary similarity detection separately.

AdGuard DNS vs AdGuard Home: 7 Smart Differences

This comparison explains managed filtering versus self-hosting, which matters before moving your network-wide DNS policy.

Troubleshooting: when the result looks wrong

No new queries appear in Private DNS

Check the selected server, actual device DNS, browser Secure DNS, VPN settings, IPv6 and cached results. Request a harmless hostname and compare the timestamp. If a private server never receives the query, its typosquatting protection policy cannot act on that request.

A suspicious spelling is not blocked

Check the switch and server association first, then confirm the request reached that resolver. An enabled classifier can still miss a particular variation. Do not conclude that every detected domain is malicious or every undetected domain is safe.

An expected website stopped working

Identify the blocking rule if possible, verify the site independently, and add only a narrow allow rule when justified. Other application or browser filters might also block the hostname. A DNS exception does not override every security layer, nor should it.

Is DNSSEC the same as typo detection?

No. DNSSEC helps validate certain DNS responses and reduces some forms of answer forgery when properly implemented. A lookalike domain can be validly registered and have valid DNSSEC and HTTPS. Typosquatting protection addresses misleading names, not whether signed DNS data came from the legitimate owner of an unrelated brand.

When your concern is interception rather than a lookalike domain, consult the separate HackersGhost guide on network interception. It discusses a different threat model that a DNS similarity setting does not solve.

Man in the Middle Attacks: 7 Alarming Network Tricks

Differentiate forged destinations from intercepted traffic so you select the right defensive control.

Does a paid DNS plan make sense for this protection?

A managed subscription becomes valuable when you need central security settings, per-device visibility, query history and safe exceptions across multiple devices. If you only want ordinary DNS filtering, a free public resolver or another trusted provider might be sufficient. The purchase decision should follow a successful resolver-path check, not precede it.

If Private DNS suits your setup, the AdGuard DNS one-year offer can be an option. The exclusive HackersGhost code applies to eligible one-year plans. The final price is determined at checkout and public promotions can differ. Nothing in this guide depends on a short-lived sale.

Eligible one-year plans only. Code HACKERSGHOST20 applies automatically; check the final price before payment. Public promotions may offer similar or different pricing.

HackersGhost Final Note: I would rather document one real DNS policy decision than celebrate a hundred unverified warnings. Typosquatting protection earns its place when it reduces exposure and you can explain the limits. The resolver is a gatekeeper, not a psychic.

Single retro pop-art question mark for frequently asked questions about AdGuard DNS

Typosquatting protection FAQ

Can AdGuard DNS block every fake website?

Is typosquatting protection enabled automatically?

Are homograph attacks and typosquatting identical?

Can typosquatting protection block legitimate sites?

Does a VPN replace DNS phishing protection?

Is a domain starting with xn-- dangerous?

Can I test with a real phishing site?

Final verdict: enable, verify, and keep your guard up

AdGuard DNS v2.21 brought similarity-based typosquatting protection, and v2.22 added a separate switch targeting certain IDN homograph attacks. Those features can reduce exposure to deceptive hostnames, especially when combined with appropriate security lists. They will not resolve every phishing threat, and a configuration that a device bypasses offers that device no protection.

Your first action is simple: choose the intended Private DNS server, confirm both security controls, and verify that a harmless query from the target device reaches the correct Query log. Only then investigate blocks or add exceptions. Keep phishing-resistant sign-in and independent verification habits in place. Knowing which door you intended to open is still better than trusting a sign painted on the wrong one.

VPN & Network Infrastructure Cluster

AdGuard Ecosystem

ⓘ

Some links in this article are affiliate links. If you use them, I may earn a small commission — at no extra cost to you. I only recommend tools I’ve actually tested inside my own cybersecurity lab. Read the full disclaimer.

In many cases, these links unlock better deals than you’ll find on your own.
No paid reviews. No sponsored opinions. Just real testing and real setups.

If you decide to use them, you’re not just getting a discount — you’re helping keep this lab running.

Leave a Reply

Your email address will not be published. Required fields are marked *