Skip to content

Home/Digest/Past issues

Daily digest,

Malicious Python Package Compromises Cloud Credentials via PyPI (+3 more)

Vulnerabilities

Emerging AI Threats: ARTEX Exploits Vulnerabilities in Cloud Environments

ARTEX, an AI-powered penetration testing platform, is being weaponized by nation-state actors and cybercriminal groups for automated exploitation of vulnerabilities. It has been linked to campaigns involving espionage and supply chain compromises, notably targeting Azure environments. The framework automates reconnaissance and credential harvesting, making traditional defenses less effective. SentinelOne's analysis highlights its use in operations associated with China, where it has been observed executing complex attack chains. Security professionals are advised to rotate exposed secrets and audit web APIs to mitigate risks. The current status indicates that organizations must enhance their defenses against these AI-driven threats.

Malware · 2 sources · score 70 · BloodAlchemy

Threat actors and malware

Malicious Python Package Compromises Cloud Credentials via PyPI

On November 17, 2026, two malicious versions (1.26.1 and 1.26.2) of Microsoft's agent-framework package were published on PyPI, containing a credential stealer that targets Azure, Azure DevOps, and GitHub Actions secrets. The malware executes on Python startup through .pth files, collecting sensitive information and sending it to an attacker-controlled endpoint. Affected users include those who installed these versions directly or indirectly during a 45-minute window. The package sees approximately 190,000 downloads monthly, raising concerns about widespread exposure. Users are advised to pin to version 1.26.0, remove the malicious versions, and rotate any potentially compromised credentials. The incident highlights ongoing risks associated with supply-chain attacks in software dependencies.

Malware · 2 sources · score 72 · Agent-framework

AI-Driven Mobile Malware Threatens Corporate Security

Recent reports reveal a surge in AI-assisted mobile malware targeting corporate environments. Attackers are exploiting counterfeit mobile applications to harvest credentials and sensitive data, impacting enterprise security. The Zimperium 2026 zLabs Year in Review highlights that generative AI is being used to create high-fidelity clones of legitimate apps, facilitating phishing and financial fraud. The report identified 608 websites distributing Coruna-family code, which poses browser-based threats to iOS devices. As mobile workforces grow, the risk of mobile impersonation and credential theft increases. Companies are urged to enhance mobile security measures as traditional defenses become inadequate against these evolving threats. The Cybersecurity and Infrastructure Security Agency (CISA) has issued advisories on these tactics, which include multi-stage delivery methods and zero-click exploits. The situation remains critical as attackers refine their strategies.

Malware · 2 sources · score 64 · Arsink RAT, Coruna-family, Mantax Otax, PixRevolution, RecruitTrap

TheMailer Phishing Kit Leverages AI for Targeted Campaigns

TheMailer is a new phishing kit identified by Varonis that utilizes AI models to automate and personalize phishing attacks. It allows operators to create tailored email lures that can bypass spam filters, targeting tens of thousands of victims at once. The kit can generate unique emails for each recipient, enhancing the effectiveness of bulk campaigns. Operators purchase a license for $3,000 to $6,000 in cryptocurrency and supply their own API keys for AI models like Claude and ChatGPT. The kit's capabilities include generating lures, adjusting content for spam filters, and managing replies. While the success rates of these attacks are unclear, Varonis suggests a significant customer base for TheMailer. This development indicates a trend towards more sophisticated phishing tools in the cybercriminal ecosystem.

Phishing · 3 sources · score 51

New on leak sites

53 victims listed on ransomware leak sites by 18 groups in the 24 hours before this issue. The most active:

Also moving

  • Malware: Cobalt Strike, Gootloader, Agent-framework, Aoki, Arsink RAT
  • CVEs: CVE-2026-109618, CVE-2026-109619, CVE-2026-114938, CVE-2016-4570, CVE-2016-4571
  • Vulnerabilities: Java4Shell, WordPress Flaw, Argument Injection, BadHost
  • Campaigns: Operation Epic Fury, Anonymous Hacktivist Campaign, Belém Declaration On Information Integrity On Climate Change, Be Ready, BioCatch Trust New Zealand
  • Ransomware groups: Akira
  • APT groups: APT28, APT41, Bronze Sandcastle, Cosmic Leopard

Get the next one by email

The digest is free and arrives every morning. One click to leave.

Subscribe to the digest

A free account turns the digest into a personal watchlist: choose what you want to follow.