Computer Software/Engineering
Breach intelligence, attack campaigns, and threat reports targeting the Computer Software/Engineering sector.
Explore Other Sectors
Computer Software/Engineering Threat Reports
PyTorch Lightning Supply Chain Attack: What You Need to Know
In April 2026, threat actors compromised the PyTorch Lightning package by publishing malicious versions 2.6.2 and 2.6.3 on the Python Package Index (PyPI). These versions contained obfuscated JavaScript payloads that executed upon import, leading to the theft of credentials, authentication tokens, and cloud secrets. The attack also attempted to poison GitHub repositories by creating public repositories with names like 'EveryBoiWeBuildIsaWormBoi'. The malicious versions were quickly identified and removed from PyPI, and developers were advised to downgrade to version 2.6.1 and rotate any potentially exposed credentials. ([semgrep.dev](https://semgrep.dev/blog/2026/malicious-dependency-in-pytorch-lightning-used-for-ai-training?utm_source=openai)) This incident underscores the growing trend of supply chain attacks targeting widely-used open-source packages. The rapid detection and response highlight the importance of vigilant monitoring and prompt action in mitigating such threats. Organizations are reminded to regularly audit their dependencies and implement robust security practices to protect against similar attacks.
4 months ago
Kill Chain
AI Uncovers Critical Vulnerabilities in OpenEMR EHR Platform
In April 2026, an AI-driven analysis by cybersecurity firm Aisle uncovered 38 previously unknown vulnerabilities in OpenEMR, an open-source electronic health record platform utilized by over 100,000 healthcare providers globally. These vulnerabilities, ranging from medium to critical severity, included issues like missing authorization checks, cross-site scripting (XSS), SQL injection, path traversal, and session-related flaws. Exploitation of these vulnerabilities could have led to full database compromises, large-scale exfiltration of protected health information (PHI), and remote code execution on servers. ([darkreading.com](https://www.darkreading.com/vulnerabilities-threats/ai-finds-38-security-flaws-openemr?utm_source=openai)) The rapid identification and remediation of these flaws underscore the transformative impact of AI in vulnerability research, significantly reducing the time required for such analyses. However, this also highlights the growing challenge for security teams to triage and address the increasing volume of discovered vulnerabilities, emphasizing the need for robust and proactive cybersecurity measures in the healthcare sector.
4 months ago
Kill Chain
Critical RCE Vulnerability CVE-2026-3854 in GitHub Enterprise Server
In March 2026, a critical remote code execution (RCE) vulnerability, CVE-2026-3854, was identified in GitHub Enterprise Server. This flaw allowed attackers with push access to a repository to execute arbitrary code on the server by exploiting improperly sanitized push option values during git push operations. The vulnerability was reported by cloud security firm Wiz through GitHub's bug bounty program and was promptly addressed by GitHub, with patches released for affected versions. The discovery of CVE-2026-3854 underscores the evolving landscape of cybersecurity threats, particularly the role of AI in identifying vulnerabilities within closed-source binaries. This incident highlights the necessity for organizations to adopt proactive security measures and stay vigilant against emerging attack vectors facilitated by advanced technologies.
4 months ago
Kill Chain
Claude Mythos AI Exposes Critical Vulnerabilities in Japan's Financial Systems
In April 2026, Anthropic's advanced AI model, Claude Mythos, demonstrated the capability to autonomously identify and exploit previously unknown vulnerabilities across major operating systems and web browsers. This revelation prompted Japan's financial authorities, including the Financial Services Agency and the Bank of Japan, to establish a task force aimed at mitigating potential cybersecurity threats to the nation's financial infrastructure. The task force's formation underscores the urgency of addressing AI-driven cyber risks in a sector heavily reliant on interconnected and legacy systems. The emergence of AI models like Claude Mythos signifies a paradigm shift in cybersecurity, where the speed and sophistication of potential attacks could outpace traditional defense mechanisms. Financial institutions worldwide are now compelled to reassess and fortify their security postures to counteract the evolving threat landscape posed by advanced AI capabilities.
4 months ago
Kill Chain
EtherRAT Campaign: A New Era of Malware Distribution via GitHub and Ethereum
In March 2026, a sophisticated cyber campaign was identified targeting enterprise administrators, DevOps engineers, and security analysts. The attackers employed SEO poisoning to manipulate search engine results, leading victims to GitHub repositories that impersonated legitimate administrative tools. These repositories hosted malicious MSI installers, which, upon execution, deployed EtherRAT—a Node.js-based backdoor. Notably, EtherRAT utilized Ethereum smart contracts to dynamically resolve command-and-control (C2) addresses, enhancing the malware's resilience and evasion capabilities. This incident underscores a strategic shift in cyberattack methodologies, combining social engineering with decentralized technologies to evade detection and maintain persistence. The use of blockchain for C2 infrastructure highlights the evolving tactics of threat actors, necessitating adaptive defense strategies to counter such innovative threats.
4 months ago
Kill Chain
Google Patches Critical RCE Vulnerability in Gemini CLI
In April 2026, Google addressed a critical security vulnerability in the Gemini CLI, specifically affecting the "@google/gemini-cli" npm package and the "google-github-actions/run-gemini-cli" GitHub Actions workflow. This flaw, assigned a CVSS score of 10.0, allowed unprivileged external attackers to execute arbitrary commands on host systems by injecting malicious content into Gemini configuration files. The vulnerability was particularly concerning in Continuous Integration (CI) environments where Gemini CLI operated in headless mode, automatically trusting workspace folders and potentially leading to remote code execution via malicious environment variables in the local .gemini/ directory. ([thehackernews.com](https://thehackernews.com/2026/04/google-fixes-cvss-10-gemini-cli-ci-rce.html?utm_source=openai)) The incident underscores the critical importance of securing CI/CD pipelines against supply chain attacks. As organizations increasingly rely on automated workflows, ensuring that tools like Gemini CLI do not implicitly trust unverified inputs is essential to prevent potential exploitation and maintain the integrity of development environments.
4 months ago
Kill Chain
DEEP#DOOR: Unveiling the Python Backdoor Exploiting Tunneling Services
In April 2026, cybersecurity researchers identified a sophisticated Python-based backdoor framework named DEEP#DOOR targeting Windows systems. The attack initiates with an obfuscated batch script that disables Windows security features and extracts an embedded Python payload, establishing persistence through multiple mechanisms such as startup folder entries, registry run keys, and scheduled tasks. The malware communicates with attacker infrastructure via a public TCP tunneling service, enabling remote command execution and extensive surveillance capabilities, including keylogging, screenshot capture, and credential harvesting from browsers and cloud services. DEEP#DOOR employs advanced evasion techniques, including sandbox and virtual machine detection, to avoid detection and complicate incident response efforts. This incident underscores the evolving sophistication of threat actors who leverage fileless, script-driven intrusion frameworks that utilize native system components and interpreted languages like Python. The use of public tunneling services for command-and-control communications highlights a trend towards minimizing forensic footprints and blending malicious traffic with legitimate network activity, posing significant challenges for traditional detection methods.
4 months ago
Kill Chain
Bishop Fox Unveils AIMap: A New Tool for Securing AI Agent Infrastructures
On April 30, 2026, Bishop Fox introduced AIMap, an open-source tool designed to help organizations discover, analyze, and test their exposed AI agent infrastructure. AIMap enables defenders to identify internet-exposed AI systems, assess their risk levels, and conduct controlled security testing to understand and mitigate real-world attack surfaces. The tool addresses vulnerabilities such as unauthenticated access, tool abuse, and prompt leakage, which are increasingly exploited by attackers. The release of AIMap is particularly relevant as AI systems become more integrated into organizational operations, presenting new attack vectors. By providing visibility into AI agent infrastructures, AIMap empowers organizations to proactively secure their AI deployments against emerging threats.
4 months ago
Kill Chain
RedTail Malware's Exploitation of PHP Vulnerability CVE-2024-4577: A 2026 Cybersecurity Threat
In April 2026, a significant cybersecurity incident was documented involving the exploitation of the PHP-CGI vulnerability CVE-2024-4577 by the RedTail cryptomining malware. Attackers utilized the 'libredtail-http' User-Agent to perform HTTP POST actions, leading to unauthorized command execution and deployment of the RedTail malware. This campaign targeted various systems globally, aiming to hijack computing resources for illicit cryptocurrency mining. The rapid exploitation of this vulnerability underscores the critical need for timely patching and robust security measures to prevent such attacks. Organizations are advised to update their PHP installations promptly and monitor network traffic for indicators of compromise associated with the 'libredtail-http' User-Agent.
4 months ago
Kill Chain
Anthropic's Claude Mythos AI Model: A Double-Edged Sword in Cybersecurity
In April 2026, Anthropic unveiled 'Claude Mythos Preview,' an advanced AI model capable of autonomously identifying and exploiting thousands of zero-day vulnerabilities across major operating systems and web browsers. Due to the potential misuse of its capabilities, Anthropic restricted its release, collaborating with over 50 major organizations, including Amazon, Google, Microsoft, and the U.S. government, under 'Project Glasswing' to responsibly address these vulnerabilities. This development underscores the dual-edged nature of AI advancements in cybersecurity, highlighting the need for stringent controls and ethical considerations in deploying such powerful technologies. The emergence of AI agents like Mythos signifies a paradigm shift in cybersecurity, where autonomous systems can both fortify and threaten digital infrastructures. Organizations must adapt by implementing robust identity and access management frameworks tailored for AI entities to mitigate risks associated with their deployment.
4 months ago
Kill Chain
Vercel Breach 2026: Lessons in Third-Party AI Tool Security
In April 2026, Vercel, a cloud development platform, experienced a security breach originating from a compromised third-party AI tool, Context.ai. An attacker exploited this tool to access a Vercel employee's Google Workspace account, subsequently infiltrating Vercel's internal systems. This led to unauthorized access to non-sensitive environment variables, including API keys and database credentials. The breach was traced back to the Lumma Stealer malware, which had infected a Context.ai employee's workstation in February 2026. The malware harvested credentials, enabling the attacker to pivot into Vercel's infrastructure. Vercel has since notified affected customers and recommended immediate credential rotation and enhanced security measures. This incident underscores the escalating risks associated with third-party integrations and the necessity for stringent access controls and continuous monitoring of OAuth permissions. Organizations are urged to reassess their security postures concerning third-party tools to mitigate potential supply chain vulnerabilities.
4 months ago
Kill Chain
GitHub's Swift Response to CVE-2026-3854: Securing Millions of Repositories
In March 2026, GitHub addressed a critical remote code execution vulnerability, identified as CVE-2026-3854, which affected GitHub.com and GitHub Enterprise Server. The flaw allowed attackers with push access to a repository to execute arbitrary code on the server by exploiting improperly sanitized user-supplied push options during git push operations. This vulnerability posed a significant risk, potentially exposing millions of private repositories to unauthorized access. GitHub promptly patched the issue within hours of its discovery, and no evidence of exploitation prior to the fix was found. The rapid identification and remediation of CVE-2026-3854 underscore the importance of proactive vulnerability management in safeguarding critical code repositories. This incident highlights the necessity for organizations to maintain vigilant security practices and promptly apply patches to mitigate emerging threats.
4 months ago
Kill Chain
Stop Active Cloud Data Exfiltration
Aviatrix Rapid Containment helps teams instantly identify what data is leaving the environment, from which workload, and where it’s going — during an active breach.
Looking for threats in a different sector?
Browse All Threat Reports