737 Chrome VPN Extensions Exposed: How Fake VPN Add-ons Redirected Browser Traffic
Researchers linked hundreds of Chrome VPN and proxy extensions to a coordinated operation involving brand impersonation, browser-wide SOCKS5 proxying and more than 75,000 installs.
Socket researchers have linked 737 free Chrome VPN and proxy extensions to one coordinated campaign. The extensions accumulated a reported 75,486 installs and were published through at least 40 Chrome Web Store developer accounts.
The most important finding is not that every extension was proven to steal passwords or record users' traffic. It is that hundreds of the extensions were capable of redirecting a user's browser session through SOCKS5 proxy infrastructure associated with the campaign, while many also copied the names or branding of established privacy services.
Socket analysed code from 525 extensions. Its campaign-wide code statistics use a bulk-retrieved set of 522 packages, and 520 of those 522 configured Chrome to use the same underlying SOCKS5 infrastructure. The remaining extensions were either analysed separately or had already disappeared from the store before their code could be collected.
VPN and proxy extensions connected by code, infrastructure and publisher evidence.
Socket's sum of Chrome Web Store install-count buckets across the full set.
Extensions that imitated 66 established VPN and privacy brands.
The distinction matters: Socket says the extensions placed the proxy operator in a position to observe browser traffic, but the researchers did not claim to have proved what the proxy servers logged, retained or transmitted.
What Researchers Found in the Chrome VPN Extension Campaign
Socket published its expanded investigation on 11 August 2026 after examining a much larger cluster than earlier reporting had identified. The campaign targeted mainly Russian-speaking users looking for ways to access services blocked inside Russia, including Instagram, ChatGPT and YouTube.
Of the 734 extensions in Socket's original source dataset, 690 showed at least one sign of targeting Russian-speaking users. The researchers also found common infrastructure, repeated code patterns, shared build artefacts and links to a commercial VPN operation using the Myxa VPN name.
Palo Alto Networks publicly described a smaller fake-VPN extension farm using hardcoded SOCKS5 proxies and impersonating several VPN brands.
Socket published a broader analysis linking 737 extensions and more than 75,000 reported installs to the same campaign.
The practical advice remains to verify publishers carefully and remove any extension identified as part of the campaign.
A VPN extension is given unusually sensitive control because redirecting network traffic is part of what users expect it to do. That makes fake privacy tools especially dangerous: malicious-looking behaviour can be disguised as normal VPN functionality.
How the Extensions Redirected Browser Traffic
Chrome extensions can use the browser's proxy API to change how Chrome reaches websites. In a legitimate browser VPN or proxy extension, that can be the intended feature. In this campaign, Socket found that 520 of the 522 packages used for its main code-level analysis configured a fixed SOCKS5 proxy on port 1082.
The bypass list contained only local loopback addresses. Once the user clicked connect, ordinary browser destinations were therefore sent through the configured proxy instead of connecting directly.
The user installs a VPN or proxy extension
The listing may use generic privacy language or imitate a familiar VPN brand.
The extension requests proxy control
The permission looks plausible for a VPN-style browser extension because proxy configuration is central to the advertised function.
Chrome is pointed at a SOCKS5 relay
Socket found hundreds of packages using shared proxy infrastructure, with many hardcoding IP addresses from the same pool.
Browser requests pass through the proxy
The proxy can see connection metadata and any unencrypted HTTP content passing through it.
What could the proxy see?
Socket says the proxy infrastructure was positioned to see the user's source IP address, destination information and TLS Server Name Indication values for HTTPS connections. For plain HTTP traffic, the proxy could also read request contents, including credentials submitted to an unencrypted site.
HTTPS still protects the encrypted contents of a normal TLS connection from a simple SOCKS5 relay. That is why it would be inaccurate to say the research proved that every HTTPS password, message or page body was readable.
This was browser proxying, not a full-device VPN tunnel. The extension changed Chrome's network path. It did not automatically protect or redirect traffic from every other application running on the device.
DNS-over-HTTPS made some infrastructure harder to block
Socket found 104 extensions that used Cloudflare or Google DNS-over-HTTPS endpoints to resolve their own proxy hostnames and then supplied Chrome with the resulting raw IP address. The technique meant the local device did not need to make an ordinary plaintext DNS lookup for the campaign's proxy domain.
274 Extensions Impersonated 66 VPN and Privacy Brands
The brand-abuse element is one of the strongest reasons to treat the campaign as more than a questionable proxy service. Socket counted 274 extensions impersonating 66 established VPN and privacy brands, representing 38,140 reported installs.
Names copied or imitated included Proton VPN, NordVPN, Surfshark, AdGuard VPN, Browsec, ExpressVPN, CyberGhost, Windscribe, TunnelBear, Cloudflare's 1.1.1.1 and Google's Outline. The researchers also highlighted AmneziaVPN and AntiZapret because they are specifically associated with censorship circumvention for Russian users.
| Signal | Why it should concern users |
|---|---|
| Copied VPN brand name or logo | A familiar name does not prove that the Chrome Web Store publisher is the real provider. |
| Publisher identity does not match the provider | The extension may be unaffiliated even if its listing looks convincing. |
| Proxy permission | This permission can redirect essentially all Chrome browsing traffic while the extension is connected. |
| Unexpected redirects after installation | Socket found first-install behaviour that pushed users towards paid subscription pages. |
| Claims that cannot be independently verified | Researchers found premium server locations that did not resolve during testing. |
Socket also found evidence suggesting attempts to make the extension estate harder to review and disrupt. Some extensions resolved campaign domains remotely, some could rediscover configuration after redirects, and 49 extensions across 18 publisher accounts received post-approval code changes in the dataset examined.
Does This Mean Chrome VPN Extensions Are Unsafe?
No. The proxy mechanism itself is not evidence of malicious behaviour. A legitimate browser VPN or proxy extension may need to use Chrome's proxy controls to provide its service.
The problem in this case is the wider pattern: large-scale impersonation, shared undisclosed infrastructure, misleading review justifications described by Socket, premium locations that did not resolve during the investigation, and code changes after store approval.
A Chrome Web Store listing is not the same as provider verification
Before installing a VPN extension, start from the VPN provider's own verified website and follow its link to the browser store. Then compare the developer or publisher identity, website and privacy information rather than relying on the extension name alone.
Users should also understand the difference between a VPN browser extension and a full VPN application. A browser extension generally protects or proxies browser traffic only. A full VPN app normally creates an operating-system-level tunnel that can cover other applications as well.
What to Do If You Installed a Suspicious Chrome VPN Extension
- Remove the extension: in Chrome, open Extensions and then Manage extensions, find the add-on and select Remove.
- Check Chrome's proxy state: Socket recommends confirming that the browser or system proxy configuration has returned to normal after removal.
- Review what you did while connected: treat browsing destinations during that period as potentially visible to the proxy operator.
- Change credentials entered over plain HTTP: if you submitted a password or other sensitive information to a non-HTTPS page while the extension was connected, change it.
- Verify the real provider: if you still want a VPN extension, reach its store page through the provider's official website rather than searching by name alone.
- Review other extensions: remove unfamiliar add-ons and pay particular attention to extensions with powerful network or site-access permissions.
How to check whether a VPN extension is genuine
1. Start with the provider's website
Use the VPN company's verified domain and follow its browser-extension link. Do not assume the first similarly named Web Store result is official.
2. Check the publisher and linked website
The store publisher, support page and privacy-policy domain should make sense for the real company. Small spelling differences and unrelated domains are warning signs.
3. Read the permission prompt
A VPN extension may legitimately require proxy access, but that permission is powerful. Only grant it when you are confident about who operates the extension.
4. Do not rely on ratings alone
Socket reported coordinated review activity in this campaign, so a high rating should not override doubts about publisher identity or unexpected behaviour.
What This Means
The 737-extension campaign is a useful reminder that a privacy label is not a security guarantee. Browser VPN extensions occupy a position of trust because routing traffic is their stated purpose, and an impersonator can abuse that expectation without looking like conventional malware.
The safest approach is to verify the publisher through the provider's official website, understand exactly what traffic a browser extension covers, and remove any extension connected with this campaign. For the affected set, the central risk established by the research is traffic redirection through infrastructure the user did not knowingly choose or verify.
Frequently Asked Questions
Were all 737 Chrome VPN extensions proven to steal users' data?
No. Socket linked 737 extensions to the campaign and established widespread proxy redirection and impersonation. It explicitly said its research did not establish what the proxy servers retained or transmitted.
How many extensions actually redirected browser traffic?
Socket's main code-level figures use 522 bulk-retrieved packages. Of those, 520 configured a proxy. Socket analysed 525 extensions in total, with three additional packages examined separately.
Which well-known VPN brands were impersonated?
The list included Proton VPN, NordVPN, Surfshark, ExpressVPN, CyberGhost, Windscribe, TunnelBear, AdGuard VPN, Browsec and others. Cloudflare's 1.1.1.1 and Google's Outline were also impersonated.
Is a SOCKS5 proxy the same as a VPN?
No. A SOCKS5 proxy can relay browser connections, but it does not by itself create the same encrypted, operating-system-level tunnel normally associated with a full VPN application.
What should I do if I installed one of the suspicious extensions?
Remove it, confirm your proxy settings are normal, review sensitive activity performed while it was connected and change credentials submitted over non-HTTPS connections.
