Researchers have identified malicious packages targeting popular developer ecosystems including VS Code extensions, Go modules, npm packages, and Rust crates. These malicious packages are designed to steal sensitive developer data, potentially including credentials, environment variables, and source code. The threat is significant due to the widespread use of these package managers and development tools across the software industry. Exploitation does not require advanced user interaction beyond installing or updating packages, making it relatively easy to execute. Although no known exploits are currently active in the wild, the potential for data exfiltration and supply chain compromise is high. European organizations relying heavily on these development environments are at risk, especially those in software development, technology services, and critical infrastructure sectors. Mitigation requires strict vetting of third-party packages, use of code signing, dependency auditing, and monitoring for unusual network activity from development environments. Countries with large software industries and strong developer communities such as Germany, France, the UK, and the Netherlands are most likely to be affected. Given the high potential impact on confidentiality and integrity, ease of exploitation, and broad scope, this threat is assessed as high severity. Defenders should prioritize supply chain security controls and developer awareness to reduce risk.
We break down a new infostealer attack that combines the ClickFix technique with a shared chat containing malicious user guides on the official ChatGPT website.
JS#SMUGGLER is a threat actor or campaign leveraging compromised legitimate websites to deliver the NetSupport Remote Access Trojan (RAT). The attack involves injecting malicious JavaScript payloads into trusted sites, which then silently deploy NetSupport RAT onto victims' systems. This RAT enables attackers to gain persistent remote access, potentially leading to data theft, espionage, or further network compromise. The threat is considered high severity due to the use of trusted sites for delivery, the capabilities of NetSupport RAT, and the stealthy infection vector. European organizations relying on web traffic from compromised sites are at risk, especially those with remote workforce or critical infrastructure. Mitigation requires proactive monitoring of web assets, enhanced endpoint detection, and network segmentation to limit RAT impact. Countries with high internet usage, significant remote workforces, and strategic industries are more likely targets. Given the ease of exploitation via compromised sites and the high impact on confidentiality and integrity, the threat is assessed as high severity. Defenders should prioritize detection of unusual RAT activity and patch any web application vulnerabilities that could lead to site compromise.
A malicious campaign is targeting indie game platforms Itch.io and Patreon by posting fake update links in comments, which lead to downloads of LummaStealer malware. This malware uses advanced anti-analysis techniques to evade detection, including checks for virtual machines, specific usernames, and malware analysis processes. The payload is delivered via a nexe-compiled JavaScript file that drops and loads a DLL variant of LummaStealer. Despite efforts to remove malicious accounts, attackers continuously create new ones, indicating an ongoing and persistent threat. The campaign primarily targets users seeking game updates, exploiting trust in indie game communities. No known exploits in the wild have been reported yet, but the malware’s stealth and persistence pose a medium-level risk. European organizations involved in gaming, digital content creation, or using these platforms could be impacted, especially those with less mature security controls. Mitigation requires targeted detection of fake update links, monitoring of platform comments, and enhanced endpoint defenses against DLL injection and obfuscated JavaScript. Countries with active indie game development and strong Patreon/Itch.
GhostPenguin is a newly discovered, undocumented Linux backdoor malware identified through AI-driven threat hunting. It is a multi-threaded C++ program that establishes remote shell access and file system manipulation capabilities over an encrypted UDP channel using RC5 encryption. The malware employs a structured handshake and thread synchronization for registration, heartbeat, and command delivery, making it stealthy and evasive. Although still in development with debug artifacts present, GhostPenguin's capabilities include remote shell, file and directory operations, and encrypted communications. No known exploits are currently in the wild, and it was found in zero-detection samples on VirusTotal. The malware targets Linux systems, which are widely used in European enterprises and infrastructure. Detection was enabled by custom YARA rules and AI profiling, highlighting the importance of advanced threat hunting techniques. The threat poses a medium severity risk due to its stealth, remote control features, and potential for lateral movement within networks.
Ransomware payments reached the highest level in 2023, at $1.1 billion paid in 1,512 reported incidents. The post Ransomware Payments Surpassed $4.5 Billion: US Treasury appeared first on SecurityWeek .
A recent analysis reveals that malicious Visual Studio Code (VS Code) extensions can bypass the official VS Code Marketplace's security analysis, exploiting weaknesses in both the Marketplace and the OpenVSX alternative. Attackers can publish extensions that appear legitimate but contain harmful code, potentially compromising developer environments. This threat highlights the insecurity of OpenVSX, an open-source VS Code extension registry, which lacks rigorous vetting processes. European organizations relying on VS Code and OpenVSX for development tools face risks of code injection, data leakage, and supply chain attacks. Mitigations include strict extension source verification, enhanced monitoring of extension behavior, and restricting extension installation to trusted repositories. Countries with strong software development sectors and high VS Code adoption, such as Germany, France, and the UK, are particularly at risk. Given the ease of exploitation without authentication and the potential impact on confidentiality and integrity, this threat is assessed as high severity. Defenders should prioritize vetting extensions, educating developers, and employing endpoint protection to mitigate risks.
A new class of prompt injection attack vectors leveraging Model Context Protocol (MCP) sampling techniques has been identified. These attacks manipulate the input prompts to AI language models, causing them to execute unintended or malicious instructions. While no known exploits are currently active in the wild, the medium severity rating reflects the potential risks to confidentiality and integrity in AI-driven systems. The attack exploits weaknesses in how models process and sample context, potentially leading to unauthorized data disclosure or manipulation of AI outputs. European organizations using AI services that rely on prompt-based interactions or context sampling are at risk, especially those in sectors heavily adopting AI technologies. Mitigation requires careful input validation, context management, and monitoring of AI outputs for anomalies. Countries with advanced AI adoption and critical infrastructure integration, such as Germany, France, and the UK, are more likely to be targeted. Given the complexity and novelty of these attacks, the threat is assessed as medium severity due to the moderate impact and exploitation difficulty. Defenders should prioritize understanding MCP mechanisms and implementing strict prompt sanitization to reduce risk.
The Intellexa leaks have exposed zero-day vulnerabilities and an innovative attack vector using advertisements to deliver Predator spyware. This spyware is highly sophisticated, leveraging undisclosed vulnerabilities to infiltrate target systems without detection. The attack vector involves malicious ads that, when interacted with or simply viewed, can trigger the exploitation of zero-day flaws to install spyware. Although no known exploits are currently active in the wild, the potential for targeted espionage and surveillance is significant. European organizations, especially those in sensitive sectors, face elevated risks due to the spyware's stealth and advanced capabilities. Mitigation requires proactive threat hunting, enhanced monitoring of ad traffic, and collaboration with intelligence agencies to detect and block such campaigns. Countries with high adoption of affected platforms and strategic geopolitical importance are most likely targets. Given the high impact on confidentiality and the complexity of detection, this threat is assessed as high severity. Defenders should prioritize awareness, network segmentation, and zero-day vulnerability management to reduce exposure.
Predator spyware has introduced a new infection vector enabling zero-click attacks, which do not require any user interaction to compromise a target device. This advancement increases the stealth and effectiveness of the spyware, allowing attackers to silently infiltrate devices and exfiltrate sensitive data. Although no known exploits are currently observed in the wild, the high severity rating reflects the potential risk posed by this novel attack method. European organizations, especially those in sectors with high-value targets such as government, defense, and critical infrastructure, could face significant confidentiality and integrity risks. Mitigation requires proactive monitoring for unusual device behavior, applying security updates promptly, and employing advanced endpoint detection solutions capable of identifying zero-click attack patterns. Countries with high adoption of mobile and IoT technologies, and those with strategic geopolitical importance, are more likely to be targeted. Given the ease of exploitation without user interaction and the potential for broad impact, this threat is assessed as high severity. Defenders should prioritize awareness and readiness to detect and respond to such sophisticated spyware threats.
A newly reported security concern involves prompt injection attacks within GitHub Actions workflows that utilize AI agents. This threat exploits the way AI-driven automation interprets and executes prompts, potentially allowing attackers to inject malicious instructions. Although no known exploits are currently active in the wild, the medium severity rating reflects the risk of unauthorized command execution and data leakage within CI/CD pipelines. European organizations using GitHub Actions with integrated AI tools are at risk, especially those in software development and critical infrastructure sectors. Mitigation requires careful prompt sanitization, strict workflow permissions, and monitoring of AI interactions within automation. Countries with high GitHub adoption and advanced software ecosystems, such as Germany, the UK, France, and the Netherlands, are most likely to be affected. Given the potential impact on confidentiality and integrity, ease of exploitation via crafted inputs, and the broad use of GitHub Actions, the suggested severity is medium. Defenders should prioritize securing AI prompt inputs and restricting workflow access to mitigate this emerging threat.
The 25-page document outlines four principles for securely integrating AI with operational technology. The post Global Cyber Agencies Issue AI Security Guidance for Critical Infrastructure OT appeared first on SecurityWeek .
How an attacker can find out the topic of your chats with an AI assistant without hacking your computer, and what you can do to guard against this threat
Kohler's smart toilet camera, marketed as having end-to-end encryption, has been found not to provide true end-to-end encryption, potentially exposing sensitive video data to interception or unauthorized access. This vulnerability undermines user privacy and could lead to significant confidentiality breaches. Although no known exploits are currently in the wild, the high sensitivity of the data involved elevates the risk. European organizations and consumers using these devices may face privacy violations and reputational damage. Mitigation requires verifying actual encryption implementations, demanding transparency from the vendor, and applying network-level protections. Countries with higher adoption of smart home and IoT devices, such as Germany, the UK, and the Nordics, are more likely to be affected. Given the nature of the data and ease of exploitation through network interception, the severity is assessed as high. Defenders should prioritize verifying device security claims and monitoring network traffic for unauthorized access attempts.
A campaign targeting Russian corporate entities, particularly HR, payroll, and administrative departments, has been uncovered. The attack uses realistic decoy documents themed around employee bonuses and financial policies. The malware ecosystem involves a malicious LNK file leading to an implant dubbed DUPERUNNER, which then loads the AdaptixC2 Beacon to connect to the threat actor's infrastructure. The infection chain begins with a spear-phishing ZIP archive containing PDF-themed LNK files. The DUPERUNNER implant, programmed in C++, performs various functions including downloading and opening decoy PDFs, process enumeration, and shellcode injection. The final stage involves the AdaptixC2 Beacon, which communicates with the command-and-control server. The campaign, tracked as UNG0902, uses multiple malicious infrastructures and is believed to be targeting employees of various organizations.
A recently disclosed security issue involves vulnerabilities in the Picklescan mechanism used to scan PyTorch machine learning models. These bugs allow maliciously crafted PyTorch models to evade detection by security scanners and execute arbitrary code on the host system. This poses a significant risk as PyTorch is widely used in AI and machine learning workflows, including in European organizations. The threat does not currently have known exploits in the wild but is rated as high severity due to the potential for remote code execution without user interaction. Organizations relying on PyTorch models for AI development or deployment should urgently review their security posture. Mitigations include restricting model sources, isolating model execution environments, and monitoring for anomalous behavior. Countries with strong AI and tech sectors, such as Germany, France, and the UK, are likely to be most affected. Given the ease of exploitation and impact on confidentiality and integrity, the suggested severity is high.
Let’s Encrypt, a widely used certificate authority, plans to reduce the validity period of its SSL/TLS certificates from 90 days to 45 days. This change aims to improve security by limiting the window of exposure if a certificate is compromised. While this is not a vulnerability or exploit, it impacts certificate management processes and automation for organizations relying on Let’s Encrypt certificates. European organizations using Let’s Encrypt will need to adjust their renewal workflows to accommodate the shorter validity period. Failure to do so could lead to service disruptions due to expired certificates. The change does not introduce new attack vectors but increases operational demands on IT teams. Countries with high adoption of Let’s Encrypt certificates and strong digital infrastructure are more likely to be affected. This update underscores the importance of robust automation and monitoring in certificate lifecycle management. Overall, this is a security policy change rather than a direct threat or vulnerability.
Shai-Hulud V2 is an advanced malware campaign targeting the npm software supply chain, compromising over 700 npm packages and creating more than 27,000 malicious GitHub repositories. It introduces sophisticated techniques including pre-install phase execution, persistent backdoors via self-hosted GitHub Actions runners, credential harvesting and recycling across victims, and a destructive failsafe mechanism. The malware exfiltrates data through GitHub and propagates within the npm ecosystem, also exploiting Azure DevOps build agents. This supply chain attack enables persistent remote code execution and widespread credential theft without requiring user interaction. European organizations relying on npm packages and GitHub Actions for CI/CD pipelines face significant risks of data breaches, system compromise, and operational disruption. Mitigation requires enhanced supply chain security practices, strict GitHub Actions runner controls, credential hygiene, and proactive monitoring for anomalous activity. Countries with strong software development sectors and high npm usage, such as Germany, France, the UK, and the Netherlands, are particularly vulnerable. Given its broad impact and advanced persistence mechanisms, the threat severity is assessed as high.
