Mar 07, 2024NewsroomVulnerability / Web Security

Threat actors are conducting brute-force attacks against WordPress sites by leveraging malicious JavaScript injections, new findings from Sucuri reveal.

The attacks, which take the form of distributed brute-force attacks, “target WordPress websites from the browsers of completely innocent and unsuspecting site visitors,” security researcher Denis Sinegubko said.

The activity is part of a previously documented attack wave in which compromised WordPress sites were used to inject crypto drainers such as Angel Drainer directly or redirect site visitors to Web3 phishing sites containing drainer malware.

The latest iteration is notable for the fact that the injections – found on over 700 sites to date – don’t load a drainer but rather use a list of common and leaked passwords to brute-force other WordPress sites.

The attack unfolds over five stages, enabling a threat actor to take advantage of already compromised websites to launch distributed brute-force attacks against other potential victim sites –

• Obtaining a list of target WordPress sites
• Extracting real usernames of authors that post on those domains
• Inject the malicious JavaScript code to already infected WordPress sites
• Launching a distributed brute-force attack on the target sites via the browser when visitors land on the hacked sites
• Gaining unauthorized access to the target sites

“For every password in the list, the visitor’s browser sends the wp.uploadFile XML-RPC API request to upload a file with encrypted credentials that were used to authenticate this specific request,” Sinegubko explained. “If authentication succeeds, a small text file with valid credentials is created in the WordPress uploads directory.”

It’s currently not known what prompted the threat actors to switch from crypto drainers to distributed brute-force attack, although it’s believed that the change may have been driven by profit motives, as compromised WordPress sites could be monetized in various ways.

That said, crypto wallet drainers have led to losses amounting to hundreds of millions in digital assets in 2023, according to data from Scam Sniffer. The Web3 anti-scam solution provider has since revealed that drainers are exploiting the normalization process in the wallet’s EIP-712 encoding procedure to bypass security alerts.

The development comes as the DFIR report revealed that threat actors are exploiting a critical flaw in a WordPress plugin named 3DPrint Lite (CVE-2021-4436, CVSS score: 9.8) to deploy the Godzilla web shell for persistent remote access.

It also follows a new SocGholish (aka FakeUpdates) campaign targeting WordPress websites in which the JavaScript malware is distributed via modified versions of legitimate plugins that are installed by taking advantage of compromised admin credentials.

“Although there have been a variety of maliciously modified plugins and several different fake-browser update campaigns, the goal of course is always the same: To trick unsuspecting website visitors into downloading remote access trojans that will later be used as the initial point of entry for a ransomware attack,” security researcher Ben Martin said.

Jan 22, 2024NewsroomBrowser Security / Cyber Threat

Cybersecurity researchers have discovered a new Java-based “sophisticated” information stealer that uses a Discord bot to exfiltrate sensitive data from compromised hosts.

The malware, named NS-STEALER, is propagated via ZIP archives masquerading as cracked software, Trellix security researcher Gurumoorthi Ramanathan said in an analysis published last week.

The ZIP file contains within it a rogue Windows shortcut file (“Loader GAYve”), which acts as a conduit to deploy a malicious JAR file that first creates a folder called “NS-<11-digit_random_number>” to store the harvested data.

To this folder, the malware subsequently saves screenshots, cookies, credentials, and autofill data stolen from over two dozen web browsers, system information, a list of installed programs, Discord tokens, Steam and Telegram session data. The captured information is then exfiltrated to a Discord Bot channel.

“Considering the highly sophisticated function of gathering sensitive information and using X509Certificate for supporting authentication, this malware can quickly steal information from the victim systems with [Java Runtime Environment],” Ramanathan said.

“The Discord bot channel as an EventListener for receiving exfiltrated data is also cost-effective.”

The development comes as the threat actors behind the Chaes (aka Chae$) malware have released an update (version 4.1) to the information stealer with improvements to its Chronod module, which is responsible for pilfering login credentials entered in web browsers and intercepting crypto transactions.

Infection chains distributing the malware, per Morphisec, leverage legal-themed email lures written in Portuguese to deceive recipients into clicking on bogus links to deploy a malicious installer to activate Chae$ 4.1.

But in an interesting twist, the developers also left behind messages for security researcher Arnold Osipov – who has extensively analyzed Chaes in the past – expressing gratitude for helping them improve their “software” directly within the source code.