Search This Blog
Powered by Blogger.

Blog Archive

Labels

About Me

Load More
Trendy Right Now

Popular Posts

Showing posts with label GitHub. Show all posts
python
Blockchain cryptocurrency Cyber Attacks Cybersecurity GitHub Hacker JavaScript Linkedin malware python

North Korean Hacker Group Targets Cryptocurrency Developers via LinkedIn

 

A North Korean threat group known as Slow Pisces has launched a sophisticated cyberattack campaign, focusing on developers in the cryptocurrency industry through LinkedIn. Also referred to as TraderTraitor or Jade Sleet, the group impersonates recruiters offering legitimate job opportunities and coding challenges to deceive their targets. In reality, they deliver malicious Python and JavaScript code designed to compromise victims' systems.

This ongoing operation has led to massive cryptocurrency thefts. In 2023 alone, Slow Pisces was tied to cyber heists exceeding $1 billion. Notable incidents include a $1.5 billion breach at a Dubai exchange and a $308 million theft from a Japanese firm. The attackers typically initiate contact by sending PDFs containing job descriptions and later provide coding tasks hosted on GitHub. Although these repositories mimic authentic open-source projects, they are secretly altered to carry hidden malware.

As victims work on these assignments, they unknowingly execute malicious programs like RN Loader and RN Stealer on their devices. These infected projects resemble legitimate developer tools—for instance, Python repositories that claim to analyze stock market data but are actually designed to communicate with attacker-controlled servers.

The malware cleverly evades detection by using YAML deserialization techniques instead of commonly flagged functions like eval or exec. Once triggered, the loader fetches and runs additional malicious payloads directly in memory, making the infection harder to detect and eliminate.

One key malware component, RN Stealer, is built to extract sensitive information, including credentials, cloud configuration files, and SSH keys, especially from macOS systems. JavaScript-based versions of the malware behave similarly, leveraging the Embedded JavaScript templating engine to conceal harmful code. This code activates selectively based on IP addresses or browser signatures, targeting specific victims.

Forensic investigations revealed that the malware stores its code in hidden folders and uses HTTPS channels secured with custom tokens to communicate. However, experts were unable to fully recover the malicious JavaScript payload.

Both GitHub and LinkedIn have taken action against the threat.

"GitHub and LinkedIn removed these malicious accounts for violating our respective terms of service. Across our products, we use automated technology, combined with teams of investigation experts and member reporting, to combat bad actors and enforce terms of service. We continue to evolve and improve our processes and encourage our customers and members to report any suspicious activity," the companies said in a joint statement.

Given the increasing sophistication of these attacks, developers are urged to exercise caution when approached with remote job offers or coding tests. It is recommended to use robust antivirus solutions and execute unknown code within secure, sandboxed environments, particularly when working in the high-risk cryptocurrency sector.

Security experts advise using trusted integrated development environments (IDEs) equipped with built-in security features. Maintaining a vigilant and secure working setup can significantly lower the chances of falling victim to these state-sponsored cyberattacks.
Read more »
Malicious Campaign
Cyber Fraud Fake Hiring GitHub Infostealer Malicious Campaign

Developers Face a Challenge with Fake Hiring That Steals Private Data

 

Cyble threat intelligence researchers discovered a GitHub repository posing as a hiring coding challenge, tricking developers into downloading a backdoor that steals private data. The campaign employs a variety of novel approaches, including leveraging a social media profile for command and control (C&C) activities rather than C&C servers. Cyble Research and Intelligence Labs (CRIL) researchers discovered invoice-themed lures, suggesting that the campaign may be moving beyond a fake hiring challenge for developers. 

According to a blog post by Cyble researchers, 
the campaign appears to target Polish-speaking developers, and the malware exploits geofencing to restrict execution. The researchers believed that the campaign is disseminated through career sites such as LinkedIn or regional development forums. 

The fake recruitment test, dubbed "FizzBuzz," dupes users into downloading an ISO file containing a JavaScript exercise and a malicious LNK shortcut. When executed, the LNK file ("README.lnk") invokes a PowerShell script that installs a stealthy backdoor known as "FogDoor" by the researchers. 

Instead of employing C&C servers, FogDoor communicates with a social media platform using a Dead Drop Resolver (DDR) mechanism to retrieve attack directives from a profile, according to the researchers. The malware employs geofencing to limit execution to Polish victims. 

When it becomes operational, "it systematically steals browser cookies, Wi-Fi credentials, and system data, staging them for exfiltration before deleting traces," Cyble told reporters. The malware employs remote debugging to collect Chrome cookies and can work in the background, while Firefox credentials are obtained from profile directories. 

PowerShell script establishes persistence 

The PowerShell script also opens a "README.txt" file "to trick consumers into believing they are interacting with a harmless file," Cyble stated. This paper includes instructions for a code bug patch task, "making it appear innocuous while ensuring the PowerShell script executes only once on the victim's machine to carry out malicious activities." 

The PowerShell script also downloads an executable file and saves it as "SkyWatchWeather.exe" in the "C:\Users\Public\Downloads" folder. It then creates a scheduled task called "Weather Widget," which executes the downloaded file using mshta.exe and VBScript and is set to run every two minutes indefinitely. 

SkyWatchWeather.exe serves as a backdoor by utilising a social networking platform (bark.lgbt) and a temporary webhook service (webhookbin.net) as its command and control infrastructure. After authenticating its location, the malware attempts to connect to "bark.lgbt/api" in order to get further orders embedded in a social media platform's profile information. Cyble added that this setup complicates identification and removal operations.
Read more »
Vulnerabiliies
Cyber Security Cyberattacks GitHub Security Breach Supply Chain. CI/CD Vulnerabiliies

GitHub Action Security Breach Raises Concerns Over Supply Chain Risks

 


An attack of a cascading supply chain was recently triggered by the compromise of the GitHub action "reviewdog/action-setup@v1", which ultimately led to the security breach of the "tj-actions/changed-files" repository. As a result of this breach, unintended secrets about continuous integration and delivery were exposed, raising concerns about the integrity of software supply chains. 

There was a malicious code in the tj-actions/changed-files application last week, which introduced malicious code that was capable of extracting CI/CD secrets from the workflow logs and logging them within the log files. This incident affected approximately 23,000 repositories. Even though these logs were not accessible to the public, this exposure highlights significant security risks. In the case that the logs had become public, the attacker would have been able to gain unauthorized access to vital credentials.

Even though there has been an ongoing investigation into tj-actions/changed files, its developers have been unable to determine exactly how the attackers compromised GitHub's Personal Access Token (PAT) to gain access to critical data. For the unauthorized changes to be made, this token, which was used by an automated bot to modify code, appears to have played a pivotal role in the process. GitHub Actions and CI/CD pipelines need to be enhanced to prevent the spread of software supply chain vulnerabilities. This incident underscores the increasing threat of software supply chain vulnerabilities. 

A critical security breach has been identified in the widely used third-party GitHub Action, tj-actions/changed-files, that has been assigned the CVE-2025-30066 vulnerability. When a supply chain attack compromises the action that tracks file changes in pull requests and commits, it results in unauthorized disclosure of sensitive credentials since this action tracks file modifications. Among the secrets that were exposed were valid access keys, GitHub Personal Access Tokens (PATs), npm tokens, and private RSA keys. 

A security patch was implemented in version 46.0.1 as a response to the incident to mitigate the risk associated with it. As a result of an updated analysis from March 19, 2025, security researchers have suggested that this breach may have been the result of a similar compromise of another GitHub action, reviewdog/action-setup@v1, identified as CVE-2025-30154 by security researchers. Considering the timing of both incidents and the growing threat landscape surrounding software supply chains, there is a strong likelihood that there is a connection between them. 

The developments highlighted in this article underscore the importance of conducting rigorous security audits and maintaining enhanced monitoring practices within the GitHub ecosystem to prevent future threats. In the recent past, there was a security breach affecting GitHub Action tj-actions/changed-files that exposed critical security vulnerabilities in software supply chains, emphasizing the risks associated with third-party dependencies in continuous integration/continuous delivery. 

Through GitHub Actions, a widely used automation platform, developers can optimize their workflows through reusable components, allowing them to save time and money. However, due to the compromise of tj-actions/changed-files—a tool that detects changes in files in pull requests and commits—over 23,000 repositories were accessed unauthorized, resulting in the theft of sensitive workflow secrets. A security researcher first noticed unusual activity related to the repository on March 14, 2025, which led to the discovery of the breach. 

A malicious payload has been injected into CI/CD runners in an attempt to extract CI/CD runner memory, which exposed critical environment variables and workflow secrets within logs, which were discovered to have been injected by the attackers. An exploit like this could result in unauthorized access to confidential credentials, thereby posing a significant security risk to the organization. Having been provided with a critical lead by security researcher Adnan Khan, it has been confirmed that the root cause of this compromise stems from another GitHub Action called reviewdog/action-setup, which an independent organization maintains. 

The investigation revealed that the tj-actions/changed-files action was compromised because it was dependent on the tj-actions/eslint-changed-files action, which was itself dependent on the reviewdog/action-setup action. In addition to the attack on the review dog organization, multiple activities were also affected within that organization, indicating that the attack was more widespread than that. Maintainers of TJ-actions and Review Dog quickly mitigated this incident by implementing security patches and reducing further risks. 

To counteract growing threats within software supply chains, continuous security monitoring, dependency validation, and rapid mitigation strategies must be implemented to protect continuous integration/continuous delivery pipelines from future attacks. Wiz, one of the leading security firms, recommended that developers evaluate their potential exposure by performing a GitHub query to determine if any references to reviewdog/action-setup@v1 were found in their repositories. 

As part of this process, it is important to determine if any of the projects might have been compromised by the recent supply chain compromise. It would be prudent to treat the detection of double-encoded base64 payloads within workflow logs as a confirmation of the leakage of sensitive information. If this happens, immediate remediation measures are required to prevent further security incidents. 

To reduce the risks associated with compromised actions, developers are advised to remove all references to these actions across branches, remove workflow logs that might contain exposed credentials, and rotate any potentially compromised secrets so that unauthorized access cannot occur. There is a need to take proactive security measures, such as pin GitHub Actions to specific commit hashes rather than version tags to reduce the probability that similar breaches will occur in the future. Furthermore, by utilizing GitHub's allow-listing feature, we can restrict unauthorized actions and enhance the security of our repositories. 

One must respond quickly to supply chain attacks, which may have far-reaching consequences as well as leak CI/CD secrets. Immediately following the breach, organizations must take steps to contain the breach, and they must develop long-term security strategies to protect themselves against future threats as well. The companies that are potentially impacted by this GitHub Actions supply chain attack should take immediate measures to protect their systems from further harm. To effectively counteract unauthorized access and further exploitation, all exposed secrets must be rotated. This is especially true for those secrets that were used between March 14 and March 15, 2025. 

Failure to replace compromised credentials could result in further exploitation. Further, security teams need to thoroughly review CI/CD workflows, paying close attention to unexpected outputs, particularly within the section on "changed files". There is a good chance that any anomalies may indicate an unauthorized modification or possible data leak. All workflow references should be updated to point to specific commit hashes rather than mutable tags so that they can be used to enhance security and mitigate the risk of a similar incident in the future. This will reduce the risk that attackers may inject malicious code into widely used GitHub Actions in the future. 

A robust security policy is also crucial for organizations. For this reason, organizations must utilize GitHub's allow-listing feature to restrict access to unauthorized actions, and they should conduct regular security audits of their third-party dependencies before integrating them into workflows. This kind of prevention measure can greatly reduce the chances of an attack on the supply chain or an unauthorized change in the source code. As a result of the recent breach, it has been highlighted how widely used automation tools are prone to vulnerabilities, which emphasizes the need to maintain continuous security monitoring and develop proactive defence strategies. 

Although some organizations, like Coinbase, successfully mitigated the impact of this incident, it serves as a reaffirmation that all organizations should continue strengthening their security postures and remain vigilant when it comes to evolving threats in the software industry. Recent information about a security breach with GitHub Actions confirms that the threats associated with supply chain attacks are continuing to grow in the modern software development industry. It has become increasingly important for organizations to enforce strong security frameworks for the sake of preventing cyber threats by implementing continuous monitoring mechanisms, thorough dependency audits, and enhanced access controls as cyber threats become more sophisticated. 

CI/CD pipelines need to be protected against unauthorized intrusions at all costs, and this incident highlights the urgency for proactive defense strategies to prevent this type of activity. Teams can mitigate vulnerabilities and ensure their workflows are protected by adopting secure coding best practices, enforcing strict authentication policies, and utilizing GitHub's security features, if they implement secure coding practices and enforce strict authentication policies. As software supply chain security has become a world-wide concern, maintaining vigilance and immediate response to incidents is crucial to ensuring operational integrity and resilience against evolving threats in an era when it has become paramount.
Read more »
Microsoft
Cyber Attacks GitHub Infostealer Malvertising Microsoft

Microsoft Warns of Malvertising Campaign Impacting Over 1 Million Devices Worldwide

 

Microsoft has revealed details of a large-scale malvertising campaign that is believed to have impacted over one million devices worldwide as part of an opportunistic attack aimed at stealing sensitive information. 

The tech giant, which discovered the activity in early December 2024, is tracking it under the broader Storm-0408 umbrella, which refers to a group of attackers known for distributing remote access or information-stealing malware via phishing, search engine optimisation (SEO), or malvertising.

"The attack originated from illegal streaming websites embedded with malvertising redirectors, leading to an intermediary website where the user was then redirected to GitHub and two other platforms," the Microsoft Threat Intelligence team stated. "The campaign impacted a wide range of organizations and industries, including both consumer and enterprise devices, highlighting the indiscriminate nature of the attack.”

The campaign relied on GitHub to deliver initial access payloads, but payloads were also detected on Discord and Dropbox. The GitHub repositories were removed, but the number of such repositories was not disclosed. The Microsoft-owned code hosting service serves as a staging ground for dropper malware, which deploys a series of ads.

The Microsoft-owned code hosting site serves as a staging ground for dropper malware, which is in charge of launching a number of further programs such as Lumma Stealer and Doenerium, which can then collect system information. The assault also uses a sophisticated redirection chain with four to five layers, with the first redirector embedded in an iframe element on unlawful streaming websites that serve pirated content.

The entire infection sequence consists of several stages, including system discovery, information collecting, and the employment of follow-on payloads like NetSupport RAT and AutoIT scripts to assist more data theft. The remote access trojan also acts as a gateway for stealer malware. 

  • First stage: Establish a footing on target devices.
  • Second stage: system reconnaissance, collection, exfiltration, and payload delivery. 
  • Third stage: It involves command execution, payload delivery, defence evasion, persistence, command-and-control communications, and data exfiltration. 
  • Fourth stage: PowerShell script for configuring Microsoft Defender exclusions and running commands to download data from a remote server. 

Another feature of the assaults is the use of numerous PowerShell scripts to download NetSupport RAT, identify installed apps and security software, and scan for the presence of cryptocurrency wallets, which indicates possible financial data theft.

"Besides the information stealers, PowerShell, JavaScript, VBScript, and AutoIT scripts were run on the host," Microsoft said. "The threat actors incorporated use of living-off-the-land binaries and scripts (LOLBAS) like PowerShell.exe, MSBuild.exe, and RegAsm.exe for C2 and data exfiltration of user data and browser credentials.” 

The disclosure comes after Kaspersky reported that fake websites masquerading as DeepSeek and Grok artificial intelligence (AI) chatbots are being used to lure users into installing a previously unknown Python information stealer.

DeekSeek-themed decoy sites promoted by verified accounts on X (e.g., @ColeAddisonTech, @gaurdevang2, and @saduq5) have also been used to run a PowerShell script that leverages SSH to enable attackers remote access to the machine. 

"Cybercriminals use various schemes to lure victims to malicious resources,' the Russian cybersecurity company noted. "Typically, links to such sites are distributed through messengers and social networks. Attackers may also use typosquatting or purchase ad traffic to malicious sites through numerous affiliate programs.”
Read more »
Social Engineering
Bots Gaming GitHub malware Redox Stealer Social Engineering

GitHub Scam: Fake Game Mods Steal User Credentials and Data


An advanced malware campaign exploiting GitHub repositories masked as game mods (and cracked software) has been found, revealing a risky blend of automated credential harvesting and social engineering tactics. 

While going through articles on social engineering, cybersecurity expert Tim found “a relatively new scam scheme” that shocked him. “People create thousands of GitHub repositories with all sorts of things - from Roblox and Fortnite mods to "cracked" FL Studio and Photoshop,” says Tim. 

About Redox stealer

Experts have found more than 1,100 dangerous repositories spreading versions of Redox stealer, a python-based malware built to extract important data, browser cookies, gaming platform credentials, and cryptocurrency wallet keys.

When we download and run this software, the data collected from our systems is sent to some Discord server, according to Tim, where “hundreds of people crawl through the data searching for crypto wallet private keys, bank accounts and social media credentials, and even Steam and Riot Games accounts.” 

Redox Stealer Details

Redox runs via a multi-stage data harvesting process that starts with system surveillance. Talking about the technical architecture of the redox stealer, cybersecurity news portal GB Hackers says, “Initial execution triggers a globalInfo() function that collects the victim’s IP address, geolocation via the geolocation-db.com API, and Windows username using os.getenv(‘USERNAME’).”

Issues with Mitigation and GitHub’s Response

Even with GitHub’s malware detection systems, repositories stay functional because:

  1. Activities look real: Accounts with star counts and realistic commit histories escape heuristic analysis. 
  2. Encrypted Payloads: RAR passwords like “cheats4u” stop static code analysis. 
  3. Slow Takedowns: Threat actors rebuild banned repositories via automated topic permutations. 

According to GB Hackers, “The researcher’s spreadsheet of confirmed malicious repos has not yet triggered bulk takedowns, highlighting gaps in proactive monitoring.” 

Conclusion

The GitHub campaign has exposed a significant rise in exploitation of open-source forums for large-scale social engineering. “It's been a long journey and it's barely over - but I think it's more than enough to summarise and discuss the problem,” says Tim. He finds it shocking how easily the information can be accessed online for free “without Tor, without invite, without anyone's approval.”

The information is cleverly disguised as something such as “telegram bot” that sends us offers (scams) or other lucrative baits. 

Read more »
Technology
Artificial Intelligence ChatGPT CyberCrime CyberThreat Cyersecurity GitHub Shadow AI Technology

Ensuring Governance and Control Over Shadow AI

 


AI has become almost ubiquitous in software development, as a GitHub survey shows, 92 per cent of developers in the United States use artificial intelligence as part of their everyday coding. This has led many individuals to participate in what is termed “shadow AI,” which involves leveraging the technology without the knowledge or approval of their organization’s Information Technology department and/or Chief Information Security Officer (CISO). 

This has increased their productivity. In light of this, it should not come as a surprise to learn that motivated employees will seek out the technology that can maximize their value potential as well as minimize repetitive tasks that interfere with more creative, challenging endeavours. It is not uncommon for companies to be curious about new technologies, especially those that can be used to make work easier and more efficient, such as artificial intelligence (AI) and automation tools. 

Despite the increasing amount of ingenuity, some companies remain reluctant to adopt technology at their first, or even second, glances. Nevertheless, resisting change does not necessarily mean employees will stop secretly using AI in a non-technical way, especially since tools such as Microsoft Copilot, ChatGPT, and Claude make these technologies more accessible to non-technical employees.

Known as shadow AI, shadow AI is a growing phenomenon that has gained popularity across many different sectors. There is a concept known as shadow AI, which is the use of artificial intelligence tools or systems without the official approval or oversight of the organization's information technology or security department. These tools are often adopted to solve immediate problems or boost efficiency within an organization. 

If these tools are not properly governed, they can lead to data breaches, legal violations, or regulatory non-compliance, which could pose significant risks to businesses. When Shadow AI is not properly managed, it can introduce vulnerabilities into users' infrastructure that can lead to unauthorized access to sensitive data. In a world where artificial intelligence is becoming increasingly ubiquitous, organizations should take proactive measures to make sure their operations are protected. 

Shadow generative AI poses specific and substantial risks to an organization's integrity and security, and poses significant threats to both of them. A non-regulated use of artificial intelligence can lead to decisions and actions that could undermine regulatory and corporate compliance. Particularly in industries with very strict data handling protocols, such as finance and healthcare, where strict data handling protocols are essential. 

As a result of the bias inherent in the training data, generative AI models can perpetuate these biases, generate outputs that breach copyrights, or generate code that violates licensing agreements. The untested code may cause the software to become unstable or error-prone, which can increase maintenance costs and cause operational disruptions. In addition, such code may contain undetected malicious elements, which increases the risk of data breach and system downtime, as well.

It is important to recognize that the mismanagement of Artificial Intelligence interactions in customer-facing applications can result in regulatory non-compliance, reputational damage, as well as ethical concerns, particularly when the outputs adversely impact the customer experience. Consequently, organization leaders must ensure that their organizations are protected from unintended and adverse consequences when utilizing generative AI by implementing robust governance measures to mitigate these risks. 

In recent years, AI technology, including generative and conversational AI, has seen incredible growth in popularity, leading to widespread grassroots adoption of these technologies. The accessibility of consumer-facing AI tools, which require little to no technical expertise, combined with a lack of formal AI governance, has enabled employees to utilize unvetted AI solutions, The 2025 CX Trends Report highlights a 250% year-over-year increase in shadow AI usage in some industries, exposing organizations to heightened risks related to data security, compliance, and business ethics. 

There are many reasons why employees turn to shadow AI for personal or team productivity enhancement because they are dissatisfied with their existing tools, because of the ease of access, and because they want to enhance the ability to accomplish specific tasks. In the future, this gap will grow as CX Traditionalists delay the development of AI solutions due to limitations in budget, a lack of knowledge, or an inability to get internal support from their teams. 

As a result, CX Trendsetters are taking steps to address this challenge by adopting approved artificial intelligence solutions like AI agents and customer experience automation, as well as ensuring the appropriate oversight and governance are in place. Identifying AI Implementations: CISOs and security teams, must determine who will be introducing AI throughout the software development lifecycle (SDLC), assess their security expertise, and evaluate the steps taken to minimize risks associated with AI deployment. 

In training programs, it is important to raise awareness among developers of the importance and potential of AI-assisted code as well as develop their skills to address these vulnerabilities. To identify vulnerable phases of the software development life cycle, the security team needs to analyze each phase of the SDLC and identify if any are vulnerable to unauthorized uses of AI. 

Fostering a Security-First Culture: By promoting a proactive protection mindset, organizations can reduce the need for reactive fixes by emphasizing the importance of securing their systems from the onset, thereby saving time and money. In addition to encouraging developers to prioritize safety and transparency over convenience, a robust security-first culture, backed by regular training, encourages a commitment to security. 

CISOs are responsible for identifying and managing risks associated with new tools and respecting decisions made based on thorough evaluations. This approach builds trust, ensures tools are properly vetted before deployment, and safeguards the company's reputation. Incentivizing Success: There is great value in having developers who contribute to bringing AI usage into compliance with their organizations. 

For this reason, these individuals should be promoted, challenged, and given measurable benchmarks to demonstrate their security skills and practices. As organizations reward these efforts, they create a culture in which AI deployment is considered a critical, marketable skill that can be acquired and maintained. If these strategies are implemented effectively, a CISO and development teams can collaborate to manage AI risks the right way, ensuring faster, safer, and more effective software production while avoiding the pitfalls caused by shadow AI. 

As an alternative to setting up sensitive alerts to make sure that confidential data isn't accidentally leaked, it is also possible to set up tools using artificial intelligence, for example, to help detect when a model of artificial intelligence incorrectly inputs or processes personal data, financial information, or other proprietary information. 

It is possible to identify and mitigate security breaches in real-time by providing real-time alerts in real-time, and by enabling management to reduce these breaches before they escalate into a full-blown security incident, adding a layer of security protection, in this way. 

When an API strategy is executed well, it is possible to give employees the freedom to use GenAI tools productively while safeguarding the company's data, ensuring that AI usage is aligned with internal policies, and protecting the company from fraud. To increase innovation and productivity, one must strike a balance between securing control and ensuring that security is not compromised.
Read more »
Web3
Bitcoin Bull Market Crypto GitHub Technology Trading Web3

Crypto Bull Market Targeted: The Lottie-Player Security Breach


In an alarming development for the tech community, especially for those immersed in the Web3 ecosystem, a supply chain attack has targeted the popular animation library, Lottie-Player. If users fall for this prompt, it could enable attackers to drain cryptocurrency wallets. 

Given Lottie-Player's impressive tally of over 4 million downloads and its significant presence on many prominent websites for animation embedding, this incident underscores the security vulnerabilities associated with open-source libraries.

Understanding the Attack

The breach initially came to light on GitHub when a user noticed an unusual Web3 wallet prompt while integrating Lottie-Player on their website. Upon closer examination, it was discovered that versions 2.0.5, 2.0.6, and 2.0.7 of Lottie-Player, released between 8:12 PM and 9:57 PM GMT on October 30, 2024, had been tampered with and compromised.

The attack involved the introduction of malicious code into three new versions of the Lottie-Player library, a widely used tool for rendering animations on websites and applications. Threat actors infiltrated the distribution chain, embedding code designed to steal cryptocurrencies from users' wallets. This method of attack is particularly insidious because it leverages the trust developers place in the libraries they use.

The Broader Implications

Once the compromised versions were released, they were integrated into numerous high-profile projects, unknowingly exposing countless users to the threat—the malicious code activated during transactions, redirecting funds to wallets controlled by the attackers. In one notable case, a user reportedly lost 10 Bitcoin (BTC), worth hundreds of thousands of dollars, due to a phishing transaction triggered by the malicious script.

Following the discovery of the attack, the Lottie-Player team swiftly released a clean version, 2.0.8, which developers can use to replace the compromised files. To further contain the breach and limit exposure, versions 2.0.5 through 2.0.7 were promptly removed from npm and CDN providers like unpkg and jsdelivr.

Moving Forward

The attack occurred during a pivotal phase of the crypto bull market, intensifying efforts to steal increasingly valuable tokens. To mitigate risks, it's advisable to connect a wallet only for specific purposes rather than granting full-time permissions for signing transactions. Additionally, being prompted to connect a wallet immediately upon entering a website can serve as a potential warning sign.

Read more »
Sophos report
BYOVD Attack Cyber Attacks EDR GitHub Malware Attack RansomHub Sophos Sophos report

RansomHub Deploys EDRKillShifter Malware to Disable Endpoint Detection Using BYOVD Attacks

 

Sophos security researchers have identified a new malware, dubbed EDRKillShifter, used by the RansomHub ransomware group to disable Endpoint Detection and Response (EDR) systems in attacks leveraging Bring Your Own Vulnerable Driver (BYOVD) techniques. This method involves deploying a legitimate but vulnerable driver on a target device to gain escalated privileges, disable security measures, and take control of the system. 

The technique has gained popularity among various threat actors, including both financially motivated ransomware groups and state-sponsored hackers. The EDRKillShifter malware was discovered during an investigation of a ransomware incident in May 2024. The attackers tried to use this tool to disable Sophos protection on a targeted computer but were unsuccessful due to the endpoint agent’s CryptoGuard feature, which prevented the ransomware executable from running. Sophos’ investigation revealed two different malware samples, both exploiting vulnerable drivers with proof-of-concept code available on GitHub. These drivers include RentDrv2 and ThreatFireMonitor, the latter being part of an obsolete system-monitoring package. 

The malware’s loader execution process follows a three-step procedure. Initially, the attacker launches the EDRKillShifter binary with a password string to decrypt and execute an embedded resource named BIN in memory. This code then unpacks and executes the final payload, which installs and exploits a vulnerable driver to elevate privileges and disable active EDR processes. Once the driver is loaded, the malware creates a service and enters an endless loop that continuously monitors and terminates processes matching names on a hardcoded target list. Interestingly, the EDRKillShifter variants discovered were compiled on computers with Russian localization, and they exploit legitimate but vulnerable drivers, using modified proof-of-concept exploits found on GitHub. 

Sophos suspects that the attackers adapted portions of these proofs-of-concept and ported the code to the Go programming language. To mitigate such threats, Sophos advises enabling tamper protection in endpoint security products, separating user and admin privileges to prevent the loading of vulnerable drivers, and keeping systems updated. Notably, Microsoft continually de-certifies signed drivers known to have been misused in previous attacks. Last year, Sophos identified another EDR-disabling malware, AuKill, which similarly exploited a vulnerable Process Explorer driver in Medusa Locker and LockBit ransomware attacks.
Read more »
Subscribe to: Posts (Atom)