How Trivy and CheckMarx open-source solutions became the starting point for a massive TeamPCP attack on other applications, and what organizations using them should do.
PwC finds AI is amplifying speed and scale of attacks, as identity theft evolves into a cybercriminal supply chain. The post AI Speeds Attacks, But Identity Remains Cybersecurity’s Weakest Link appeared first on SecurityWeek .
The ban aligns with a White House determination that all routers produced abroad are a threat to national security. The post FCC Bans New Routers Made Outside the US Over National Security Risks appeared first on SecurityWeek .
The hackers compromised GitHub Action tags, then shifted to NPM, Docker Hub, VS Code, and PyPI, and teamed with Lapsus$. The post From Trivy to Broad OSS Compromise: TeamPCP Hits Docker Hub, VS Code, PyPI appeared first on SecurityWeek .
TeamPCP has launched a supply chain attack targeting LiteLLM, an open-source Python library used in 36% of cloud environments. Malicious versions 1.82.7 and 1.82.8 were published on PyPI, employing sophisticated techniques for payload delivery and persistence. The compromised packages exploit Python's .pth mechanism for stealthy execution across any Python process. The malware collects sensitive data including API keys, cloud credentials, and CI/CD secrets, encrypting and exfiltrating them to attacker-controlled domains. This attack follows TeamPCP's previous compromises of Aqua Security's Trivy and Checkmarx tools, highlighting an ongoing campaign against the open-source ecosystem. The incident underscores the potential for widespread impact and the need for vigilance in software supply chain security.
A malicious supply chain attack has been discovered in the Python Package Index package litellm version 1.82.8. The compromised package contains a malicious .pth file that executes automatically when the Python interpreter starts, without requiring explicit import. This file, located in site-packages/, exfiltrates sensitive information including environment variables, SSH keys, and cloud credentials to an attacker-controlled server. The payload is double base64-encoded to evade basic static analysis. PyPI administrators have quarantined the project to limit its spread. Users are advised to check for the malicious file, rotate all potentially exposed credentials, and audit their PyPI publishing process. The attack is attributed to TeamPCP and is actively exploited in the wild.
Aleksei Volkov has been sentenced to 81 months in prison for his role in Yanluowang ransomware attacks. The post US Prisons Russian Access Broker for Aiding Ransomware Attacks appeared first on SecurityWeek .
On March 19, 2026, Trivy, an open-source vulnerability scanner, was compromised in a sophisticated CI/CD supply chain attack. Threat actors, identified as TeamPCP, injected credential-stealing malware into official Trivy releases, affecting the core binary and GitHub Actions. The attack exploited mutable tags and commit identity spoofing on GitHub. The malware performed extensive credential harvesting, targeting cloud providers, Kubernetes secrets, and various application credentials. Microsoft Defender provides detection and investigation capabilities for this threat. Recommended mitigations include updating to safe versions, hardening CI/CD pipelines, enforcing least privilege, protecting secrets, and leveraging attack path analysis to reduce lateral movement risks.
CESER’s Project Armor is a five year initiative to harden the US critical energy infrastructure, including strengthening energy systems ‘to prevent and recover from wildfires and other hazards’. The post DoE Publishes 5-Year Energy Security Plan appeared first on SecurityWeek .
Agentic AI platforms are shifting from passive recommendation tools to autonomous action-takers with real system access, The post Why Agentic AI Systems Need Better Governance – Lessons from OpenClaw appeared first on SecurityWeek .
I have written about how to&#;x26;#;xc2;&#;x26;#;xa0;use IP KVMs securely, and recently, researchers at Eclypsium published yet another report on IP KVM vulnerabilities.&#;x26;#;xc2;&#;x26;#;xa0;But there is another issue I haven&#;x26;#;39;t mentioned yet with IP KVMs: rogue IP KVMs. IP KVMs are often used by criminals. For example, North Koreans used KVMs to connect remotely to laptops sent to them by their employers. The laptops were located in the US, and the North Korean workers used IP KVMs to remotely connect to them. IP KVMs could also be used to access office PCs, either to enable undetected "work from home"&#;x26;#;xc2;&#;x26;#;xa0;or by threat actors who use them to gain remote access after installing the device on site.
A new payload in the TeamPCP arsenal has been discovered, capable of wiping entire Kubernetes clusters. The script uses the same ICP canister as the CanisterWorm campaign, with consistent lateral movement via DaemonSets. However, this variant introduces a geopolitically targeted destructive payload aimed specifically at Iranian systems. The malware checks timezone and locale to identify Iranian systems, deploying privileged DaemonSets across every node in Kubernetes environments. Iranian nodes are wiped and force-rebooted, while non-Iranian nodes receive the CanisterWorm backdoor. The latest variant adds network-based lateral movement, exploiting exposed Docker APIs and using SSH for spread. This development shows TeamPCP's ability to operate at supply chain scale and their willingness to engage in destructive actions.
Mazda experienced a cyberattack resulting in the theft of sensitive employee and business partner information, including internal IDs, names, email addresses, and partner IDs from an internal management system. The breach exposed personally identifiable information (PII) that could be leveraged for identity theft, phishing, or further targeted attacks. Although no known exploits are currently active in the wild, the incident highlights vulnerabilities in Mazda's internal security posture. The attack did not disclose technical details about the attack vector or exploited vulnerabilities, limiting the ability to fully assess the attack method. The severity is assessed as medium due to the nature of the stolen data and the potential for misuse, but no direct operational disruption was reported. Organizations with similar internal management systems should review access controls and data protection measures. Countries with significant automotive industry presence and Mazda operations are at higher risk of similar attacks. Immediate mitigation should focus on enhancing internal system security, employee awareness, and monitoring for suspicious activity.
A threat actor known as TeamPCP expanded its supply chain attack from Aqua Security's Trivy to Checkmarx's AST GitHub Action. The attack, which began on March 19, 2026, involved injecting a credential-stealing payload into CI/CD pipelines across thousands of repositories. The malicious code harvested secrets from runner memory, queried cloud metadata, and exfiltrated encrypted data to typosquat domains. The Checkmarx compromise occurred approximately four days after the initial Trivy incident, using identical techniques but targeting a different action. This cascading effect demonstrates how compromised actions can be used to harvest credentials and compromise additional dependencies. Runtime detection proved effective in identifying the attack pattern across both waves, as the underlying behavior remained consistent despite changes in the delivery mechanism.
The KICS GitHub Action, an open-source infrastructure as code security scanner by Checkmarx, was compromised by TeamPCP, the group behind the recent Trivy attack. Between 12:58 and 16:50 UTC on March 23, 35 tags were hijacked, exposing users to credential-stealing malware. The attack involved staging imposter commits and updating tags using a compromised identity. The malware uses a new C2 domain, creates a fallback repository, and adds Kubernetes-focused persistence code. Additionally, two OpenVSX extensions were compromised. The payload targets cloud provider credentials and installs persistence on non-CI systems. Security teams are advised to audit workflows, search for exfiltration artifacts, and implement long-term hardening measures.
A multi-stage malware delivery campaign was uncovered, initially detected through a suspicious VBS file. The investigation revealed a complex attack infrastructure using Unicode obfuscation, PNG-based payload staging, and reflectively loaded .NET execution. The attacker utilized open directories to host multiple obfuscated VBS files, each mapping to different malware payloads including XWorm and Remcos RAT. A secondary infection vector involving a weaponized 'PDF' and batch script was also discovered. The campaign demonstrated a modular approach, allowing for payload rotation and multiple attack vectors from the same domain. This sophisticated infrastructure design enables rapid modification and expansion of available payloads without altering the initial delivery mechanism.
So, I&#;x26;#;39;ve been slow to get on the Claude Code/OpenCode/Codex/OpenClaw bandwagon, but I had some time last week so I asked Claude to review (/security-review) some of my python scripts. He found more than I&#;x26;#;39;d like to admit, so I checked in a bunch of updates. In reviewing his suggestions, he was right, I made some stupid mistakes, some of which have been sitting in there for a long time. It was nothing earth-shattering and it took almost no time for Claude, it took longer for me to read through the updates he wanted to make, figure out what he was seeing, and decide whether to accept them or tweak them. Here are a few of them.
The latest M-Trends report is based on insights from over 500,000 hours of Mandiant incident response investigations in 2025. The post M-Trends 2026: Initial Access Handoff Shrinks From Hours to 22 Seconds appeared first on SecurityWeek .
