A fraudulent website is distributing a modified portable edition of FileZilla version 3.69.5 that contains embedded malware. The archive appears legitimate and includes the authentic open-source FTP client, but attackers inserted one additional file, a rogue dynamic-link library named version.dll, before repackaging and circulating it online.
When users download this altered ZIP file, extract it, and launch filezilla.exe, Windows follows its standard DLL loading order. The operating system checks the application’s own directory before referencing system libraries stored in C:\Windows\System32. Because the malicious version.dll is placed inside the FileZilla folder, Windows loads it first. From that moment, the malicious code executes within the legitimate FileZilla process.
This method relies on a long-established Windows behavior known as DLL search order hijacking. It does not involve a vulnerability in FileZilla itself. Instead, the compromise depends on users downloading the installer from an unofficial domain such as filezilla-project[.]live, which imitates the legitimate project site. The attack spreads through deception, including lookalike domains and search engine manipulation, rather than automated self-propagation.
Archive Examination Reveals a Single Suspicious File
The compromised archive contains 918 files. Among them, 917 entries show a last-modified date of 2025-11-12, consistent with the authentic portable release of FileZilla 3.69.5. One file differs: version.dll carries a timestamp of 2026-02-03, nearly three months newer than the rest.
A genuine portable distribution of FileZilla does not include version.dll. Legitimate libraries in the package typically include files such as libfilezilla-50.dll and libfzclient-private-3-69-5.dll. The Windows Version API library normally resides inside the operating system directory and has no reason to be bundled with FileZilla. Its inclusion forms the basis of the compromise.
The SHA-256 hash of the trojanized archive is:
665cca285680df321b63ad5106b167db9169afe30c17d349d80682837edcc755
The SHA-256 hash of the malicious version.dll is:
e4c6f8ee8c946c6bd7873274e6ed9e41dec97e05890fa99c73f4309b60fd3da4
Execution Behavior Observed on a Live System
Monitoring the application with Process Monitor confirms the sideloading sequence. When filezilla.exe starts, Windows attempts to load required libraries. For files such as IPHLPAPI.DLL and POWRPROF.dll, the application directory does not contain a copy, producing “NAME NOT FOUND.” Windows then retrieves legitimate versions from the system directory.
For version.dll, however, the malicious copy is present locally. Windows maps it into memory without consulting System32. The attacker’s code now operates inside the trusted application process.
Approximately 17 milliseconds after loading, the malicious DLL attempts to locate version_original.dll in the same directory. The lookup fails. This pattern suggests DLL proxying, where attackers forward legitimate function calls to a renamed original library to preserve application stability. In this case, the renamed library was not included, which may explain abrupt application termination during testing.
FileZilla invokes LoadLibrary using only the file name rather than a full system path. While common in Windows software design, this practice enables directory-based DLL substitution.
Anti-Analysis Checks and Network Communication
Before activating its main payload, the DLL performs environmental checks. These include BIOS version inspection, system manufacturer queries, probing for VirtualBox registry keys, disk enumeration, memory allocation using write-watch techniques, and delayed execution loops. These checks aim to detect virtual machines or sandbox environments.
If the system appears genuine, the malware initiates encrypted domain resolution using DNS-over-HTTPS. It sends the following request to Cloudflare’s public resolver:
https://1.1.1.1/dns-query?name=welcome.supp0v3[.]com&type=A
Using HTTPS for DNS queries prevents traditional monitoring systems that rely on port 53 inspection from detecting the request.
After resolving the domain, the malware contacts:
https://welcome.supp0v3.com/d/callback?utm_tag=tbs2&utm_source=dll
Memory inspection revealed the embedded configuration:
{ "tag":"tbs", "referrer":"dll", "callback":"https://welcome.supp0v3.com/d/callback?utm_tag=tbs2&utm_source=dll" }
The UTM-style parameters suggest structured tracking of distribution channels.
The malware also attempts connections to 95.216.51[.]236 over TCP port 31415, a non-standard port. Ten connection attempts were recorded across two sessions, indicating retry logic designed to maintain communication.
Additional Capabilities Identified
Automated behavioral analysis indicated potential FTP credential harvesting. Because FileZilla stores connection details locally, unauthorized access could expose remote servers and hosting accounts. Other flagged behaviors included:
• Creation of suspended processes with memory injection
• Runtime .NET compilation using csc.exe
• Registry modifications consistent with persistence mechanisms
• Calls to Windows encryption-related APIs
These behaviors indicate functionality beyond simple credential theft, potentially including persistence and process manipulation.
Defensive Guidance
Users should download FileZilla exclusively from the official domain filezilla-project.org and verify the published hash values before execution. Portable installations should not contain version.dll. Its presence signals compromise.
Monitor outbound HTTPS traffic to public DNS resolvers such as 1.1.1.1 or 8.8.8.8 from non-browser applications. Review ZIP archive timestamps for inconsistencies before running software. Block the identified domains and IP address at the network perimeter if detected.
Malwarebytes reports detection and blocking of known variants of this threat.
Indicators of Compromise (IOCs)
• SHA-256 Hashes
665cca285680df321b63ad5106b167db9169afe30c17d349d80682837edcc755 FileZilla_3.69.5_win64.zip
e4c6f8ee8c946c6bd7873274e6ed9e41dec97e05890fa99c73f4309b60fd3da4 — version.dll
• Domains
filezilla-project[.]live
welcome.supp0v3[.]com
• Network Indicator
95.216.51[.]236:31415
A highly advanced phishing campaign targeted maintainers of packages on the Python Package Index (PyPI), utilizing domain confusion methods to obtain login credentials from unsuspecting developers. The campaign leverages fake emails made to copy authentic PyPI communications and send recipients to fake domains that mimic the genuine PyPI infrastructure.
The phishing operation uses meticulously drafted emails that ask users to confirm their email address for “account maintenance and security reasons,” cautioning that accounts will be suspended if not done.
These fake emails scare users, pushing them to make hasty decisions without confirming the authenticity of the communication. The phony emails redirect the victims to the malicious domain pypi-mirror.org, which mimics the genuine PyPI mirror but is not linked to the Python Software Foundation.
This phishing campaign highlights a series of attacks that have hit PyPi and similar other open-source repositories recently. Hackers have started changing domain names to avoid getting caught.
Experts at PyPI said that these campaigns are part of a larger domain-confusion attack to abuse the trust relationship inside the open-source ecosystem.
The campaign uses technical deception and social engineering. When users open the malicious links, their credentials are stolen by the hackers.
The core of this campaign depends upon domain spoofing. The fake domain uses HTTPS encoding and sophisticated web design to build its authority, which tricks users who might not pay close attention while accessing these sites. The malicious sites mimic PyPI’s login page with stark reality, such as professional logos, form elements, and styling, giving users an authentic experience.
This level of detail in the craft highlights robust planning and resource use by threat actors to increase the campaign’s effectiveness.
Users are advised to not open malicious links and pay attention while using websites, especially when putting in login details.
“If you have already clicked on the link and provided your credentials, we recommend changing your password on PyPI immediately. Inspect your account's Security History for anything unexpected. Report suspicious activity, such as potential phishing campaigns against PyPI, to security@pypi.org,” PyPI said in the blog post.
A massive cybercrime URL shortening service known as "Prolific Puma" has been uncovered by security researchers at Infoblox. The service has been used to deliver phishing attacks, scams, and malware for at least four years, and has registered thousands of domains in the U.S. top-level domain (usTLD) to facilitate its activities.
Prolific Puma works by shortening malicious URLs into shorter, more memorable links that are easier to click on. These shortened links are then distributed via email, social media, and other channels to unsuspecting victims. When a victim clicks on a shortened link, they are redirected to the malicious website.
Security researchers were able to track Prolific Puma's activity by analyzing DNS data. DNS is a system that translates domain names into IP addresses, which are the numerical addresses of websites and other devices on the internet. By analyzing DNS data, researchers were able to identify the thousands of domains that Prolific Puma was using to deliver its malicious links.
Prolific Puma's use of the usTLD is particularly noteworthy. The usTLD is one of the most trusted TLDs in the world, and many people do not suspect that a link with a usTLD domain could be malicious. This makes Prolific Puma's shortened links particularly effective at deceiving victims.
The discovery of Prolific Puma is a reminder of the importance of being vigilant when clicking on links, even if they come from seemingly trusted sources. It is also a reminder that cybercriminals are constantly developing new and sophisticated ways to attack their victims.
Here are some tips for staying safe from Prolific Puma and other malicious URL shortening services:
The security researchers who discovered Prolific Puma have contacted the United States Computer Emergency Readiness Team (US-CERT) and the Department of Homeland Security (DHS) about the service. Both agencies are working to take down Prolific Puma's infrastructure and prevent it from being used to launch further attacks.
Prolific Puma is not the first malicious URL-shortening service to be discovered. In recent years, there have been a number of other high-profile cases of cybercriminals using URL shortening services to deliver malware and phishing attacks.
The discovery of Prolific Puma is a reminder that URL shortening services can be abused for malicious purposes. Users should be cautious when clicking on shortened links, and should take steps to protect themselves from malware and phishing attacks.
A new attack targeting Secure Shell (SSH) servers has surfaced in the constantly changing world of cybersecurity. Concerningly, exploited SSH servers are now being provided as proxy pools on the dark web, which is a worrying trend. The integrity of global digital infrastructures as well as the security of sensitive data are seriously jeopardized by this trend.
The Proxyjacking Menace
Proxyjacking, as it is now termed, involves cybercriminals compromising SSH servers and selling them on the dark web as part of proxy pools. These servers are then used as a gateway for malicious activities, bypassing traditional security measures and gaining unauthorized access to networks. This technique allows attackers to conceal their true identity and location, making it difficult for cybersecurity professionals to trace and mitigate the threat.
Cloudflare, a prominent cybersecurity firm, highlights the significance of SSH in secure networking. SSH tunneling is a powerful tool for encrypting connections and safeguarding sensitive data during transmission. However, when these tunnels are breached, they become a potential point of vulnerability. Cloudflare emphasizes the need for robust security measures to protect against SSH-related threats.
SSH Tunneling and its Vulnerabilities
SSH tunneling is widely used to establish secure connections over untrusted networks. However, when improperly configured or outdated, SSH servers become susceptible to exploitation. Cybercriminals are quick to capitalize on these vulnerabilities, using compromised servers to launch attacks that can lead to data breaches, unauthorized access, and network compromise.
The exploitation of SSH servers for proxy jacking poses a significant risk to organizations and individuals alike. By leveraging these compromised servers, attackers can gain access to sensitive information, compromise critical systems, and disrupt operations. The consequences of such breaches can be severe, ranging from financial losses to reputational damage.
To defend against this emerging threat, organizations must prioritize the security of their SSH servers. Regularly updating and patching systems, implementing strong access controls, and employing advanced intrusion detection systems are essential to fortifying defenses against proxy jacking attacks. Furthermore, organizations should consider monitoring the dark web for any indications of compromised servers associated with their domains.
Proxyjacking has become more prevalent due to vulnerable SSH servers, which emphasizes the constant necessity for cybersecurity awareness. Being knowledgeable about new strategies and bolstering defenses are essential as cyber threats continue to change. Organizations may preserve their digital assets and shield themselves from the sneaky threat of proxyjacking by putting in place strong security measures and being diligent in monitoring for any breaches.
Volvo revealed in a press release that some of its research and development assets were the target of a cyberattack.
The ransomware organization Snatch reportedly released pictures of stolen Volvo papers into the darknet on November 30, according to the Swiss tech news blog INSIDE IT.
As per the company, owned by Geely of China, "Volvo Cars have learned that one of the file sources has been unlawfully acquired by a third party. The limited amount of R&D assets stolen during the hack has been confirmed by investigations so far."
An effort to sell data seized from Volvo Cars was initially discovered by French cybersecurity expert Anis Haboubi on a popular phishing site.
On December 31, 2022, a forum user going by the online alias IntelBroker reported that VOLVO CARS had been the target of a ransomware attack. He alleges that the Endurance Ransomware gang attacked the business and stole 200GB of private information that is now being peddled.
Database access, CICD access, Atlassian access, domain access, WiFi hotspots and logins, auth bearers, API access, PAC security access, employee lists, licenses, keys, and system files are all being offered by IntelBroker for $2500 in Monero, who has also shared a number of screenshots as evidence of the hack.
Based on the currently available information, the business does not believe this will affect the safety or security of its customers' cars or their personal information. Volvo, situated in Goteborg, is now investing money to electrify every vehicle in its lineup by 2030.
However, Bleeping Computer stated that the Snatch ransomware gang was claiming responsibility for the attack. A spokesman earlier told AFP that the company had not been hit by ransomware and remained in full control of its data.
On November 30, the extortion gang published a new post on their data leak website detailing how they had broken into Volvo Car Corporation's servers and taken files during the incursion. The entry included screenshots of the taken files as evidence.
Since then, Snatch has also released 35.9 MB of just what it claim are papers that were taken during the hack from Volvo's systems. Volvo refused to respond when a cybersecurity firm emailed it to ask if the screenshots published by the Snatch extortion group were actually of files stolen from its systems.