Skip to content

// Articles

389 articles — page 13 of 17

🌐 Network

Wormable Windows TCP/IP Race Condition RCE (CVE-2026-33827) — IPv6-Enabled Networks Face EternalBlue-Class Propagation Risk

A race condition in the Windows TCP/IP stack allows unauthenticated remote code execution against systems with IPv6 or IPSec enabled, demonstrated at Pwn2Own 2026 and patched in April's Patch Tuesday. The vulnerability's wormable characteristics — no user interaction, no authentication, network-adjacent propagation — place it in the same risk category as EternalBlue for environments that have not applied the April update.

#windows +7
🔑 IAM

BeigeBurrow: New Go-Based Covert C2 Agent Deployed via Active Directory RCE CVE-2026-33826

A previously undocumented post-exploitation tool named BeigeBurrow has been observed in at least two enterprise intrusions following exploitation of the Windows Active Directory RCE CVE-2026-33826. The Go-based agent uses HashiCorp's Yamux library to multiplex covert relay channels over port 443, blending into encrypted enterprise traffic. CVE-2026-33826 was patched in April Patch Tuesday; organisations that have not yet applied the patch should treat it as urgent.

#active-directory +5
🔬 Assessment

Seized Gentlemen Ransomware C2 Server Exposes 1,570 Victims — GPO Deployment Reveals Full Domain Compromise

Check Point Research's analysis of a seized SystemBC command-and-control server linked to The Gentlemen ransomware operation exposed 1,570+ victim IP addresses and documented the group's use of Group Policy Objects to deploy ransomware domain-wide. GPO-based distribution is a forensic marker that attackers achieved Domain Admin access days before encryption — defenders should treat it as an indicator of extended dwell time, not a starting point.

#ransomware +6
🗄️ Assets

Sanctioned Russian Crypto Exchange Grinex Shut Down After $13.74M Hack — Blames Western Intelligence

Grinex, a cryptocurrency exchange linked to the sanctioned Garantex operation, suspended all services after attackers drained $13.74 million in a targeted April 15 incident. The exchange blamed 'hostile state intelligence agencies,' pointing to the attack's technical sophistication. Elliptic and Chainalysis analysts have traced the funds but stop short of confirming attribution. The shutdown removes a significant node in Russia's sanctions-evasion infrastructure.

#crypto-theft +6
🛡️ SecOps

Kyber Ransomware Deploys Dual Windows and VMware ESXi Variants — Claims Post-Quantum Encryption

A new ransomware operation named Kyber is targeting enterprise Windows servers and VMware ESXi infrastructure with two distinct variants analysed by Rapid7. The Windows variant written in Rust implements genuine Kyber1024 post-quantum key encapsulation; the ESXi variant falsely markets the same capability while using ChaCha8 and RSA-4096. Both variants share Tor-based ransom infrastructure and have been deployed simultaneously on the same networks.

#ransomware +6
💻 AppSec

Marimo AI Notebook RCE CVE-2026-39987 Exploited at Scale — 662 Events in Three Days, NKAbuse Malware Deployed

CVE-2026-39987 (CVSS 9.3) in the Marimo Python notebook has been weaponised at scale, with Sysdig recording 662 exploitation events over three days and attackers completing credential theft within minutes of gaining access. The unauthenticated WebSocket RCE is being used to deploy NKAbuse, a multi-platform malware using the NKN peer-to-peer network for command and control. Upgrade to Marimo 0.23.0 immediately.

#ai-security +6
⚖️ Risk Mgmt

Anthropic's Claude Mythos AI Discovers Thousands of Zero-Days Across Every Major OS — Project Glasswing Offers Private Access

Anthropic's specialised vulnerability-hunting AI, Claude Mythos, has systematically discovered thousands of zero-day vulnerabilities across Windows, macOS, Linux, and major browsers — including a 17-year-old NFS RCE in FreeBSD and a 27-year-old OpenBSD denial-of-service. Project Glasswing provides private early access to Microsoft, Google, Apple, and select others. The implications for enterprise risk governance are immediate.

#ai-security +5
💻 AppSec

Microsoft Issues Emergency Patch for ASP.NET Core DataProtection Key Exposure — CVE-2026-40372

A critical security regression in Microsoft.AspNetCore.DataProtection (CVSS 9.1) introduced in .NET 10.0.0 causes encryption keys to leak on Linux deployments. Applications using cookie authentication, anti-forgery tokens, or TempData are at immediate risk. Update to .NET 10.0.7 now.

#asp-net-core +5
💻 AppSec

Cohere Terrarium AI Sandbox Escape — CVSS 9.3 WebAssembly Flaw Allows Root Code Execution on Host

CVE-2026-5752 (CVSS 9.3) in Cohere Terrarium allows an attacker to escape the Pyodide WebAssembly sandbox via JavaScript prototype chain traversal, achieving root code execution on the host Node.js process. Organisations running AI code execution environments should patch immediately and network-isolate these workloads.

#ai-security +6
🗄️ Assets

Everest Ransomware Claims Citizens Bank Breach — 380 GB Including 250,000 SSNs and 3.4 Million Records

The Everest ransomware group claims to have stolen 380 GB of Citizens Bank customer data via a third-party vendor, including 250,000 Social Security Numbers and 3.4 million banking records. Citizens attributes the breach to a vendor, not its core systems — but regulatory notification obligations apply regardless.

#ransomware +5
💻 AppSec

Google Antigravity AI Coding Assistant Had Two Chained Vulnerabilities — Prompt Injection to RCE and Reinstall-Surviving Backdoor

Mindgard researchers discovered two vulnerabilities in Google's Antigravity AI coding assistant: a prompt injection via the find_by_name tool that bypasses Strict Mode to achieve code execution, and a persistent backdoor via workspace trust that survives reinstallation of the IDE extension. Google has patched both; update immediately and audit workspace trust settings.

#ai-coding-assistant +5
🗄️ Assets

ShinyHunters Claims Breaches at Zara, Carnival, and 7-Eleven — Extortion Deadline Set

Prolific threat actor ShinyHunters posted simultaneous claims of data theft from Inditex/Zara, Carnival Corporation, and 7-Eleven on dark web forums on 21 April, threatening to publish stolen datasets. None of the companies has confirmed the breaches. Given ShinyHunters' track record, claims should be treated as credible pending investigation.

#data-breach +6
🔬 Assessment

CISA Adds Eight CVEs to KEV: PaperCut, JetBrains TeamCity, and Cisco SD-WAN Actively Exploited

CISA's April 20 Known Exploited Vulnerabilities addition is the largest single-day batch this month, confirming active exploitation across enterprise print management, CI/CD pipelines, content management, and Cisco SD-WAN infrastructure. The batch spans CVE publication years from 2023 to 2026, demonstrating that unpatched legacy vulnerabilities continue to be weaponised alongside newly disclosed flaws. Federal agencies face a BOD 22-01 remediation deadline, and private sector organisations should treat these as immediate prioritisation signals.

#cisa-kev +8
🌐 Network

Four Critical Cisco Flaws: Webex SSO User Impersonation (CVSS 9.8) and ISE Root Code Execution (CVSS 9.9)

Cisco patched four critical vulnerabilities across Webex Services and Identity Services Engine. CVE-2026-20184 allows unauthenticated attackers to impersonate any Webex user via crafted SSO tokens. Three ISE flaws at CVSS 9.9 let read-only admins execute arbitrary commands as root. Webex deployments with SSO require urgent manual action — Cisco's cloud fix is not sufficient without administrator intervention.

#cisco +9
🌐 Network

Public Exploit Released for Critical FortiSandbox RCE (CVE-2026-39808, CVSS 9.1) — Unauthenticated Root Access

A public proof-of-concept exploit has been released for CVE-2026-39808, a critical OS command injection vulnerability in Fortinet FortiSandbox that allows unauthenticated attackers to execute arbitrary commands as root via a single HTTP request. A companion authentication bypass flaw (CVE-2026-39813) affects the same versions. Patch to FortiSandbox 4.4.9 or 5.0.6 immediately.

#fortinet +8
🛡️ SecOps

Two Unpatched Windows Defender Zero-Days (RedSun + UnDefend) Actively Exploited — No Fix Available

A security researcher released two additional Windows Defender zero-days — RedSun and UnDefend — after Microsoft failed to patch them. RedSun exploits Defender's cloud file rollback mechanism to achieve SYSTEM privileges on all supported Windows versions. UnDefend silently prevents Defender from updating its threat signatures. Both are confirmed exploited in the wild, and neither has a patch or assigned CVE.

#windows-defender +9
💻 AppSec

Vercel Confirms Breach via Compromised AI Tool — Developer Environment Variables and Credentials Exposed

Cloud deployment platform Vercel has confirmed a breach traced to a Lumma infostealer infection at Context.ai, a third-party AI tool used by a Vercel employee. Attackers used the stolen Google Workspace OAuth access to reach Vercel's internal environments, exposing environment variables and a limited set of customer credentials. ShinyHunters is claiming responsibility and demanding $2 million for the stolen data.

#vercel +9
🗄️ Assets

McGraw Hill Confirms 13.5 Million Account Breach After ShinyHunters Exploits Salesforce Misconfiguration

Education publisher McGraw Hill has confirmed a data breach affecting 13.5 million accounts after the ShinyHunters cybercriminal group threatened to publish 45 million Salesforce records. The breach stemmed from a misconfiguration within Salesforce's environment — one McGraw Hill acknowledges is part of a broader issue affecting multiple organisations. Over 100GB of data has been publicly released.

#data-breach +7
🛡️ SecOps

Payouts King Ransomware Deploys Hidden QEMU VMs to Blind Endpoint Security — New EDR Evasion Technique

The Payouts King ransomware operation, linked to former BlackBasta affiliates, has introduced a novel EDR bypass: deploying a legitimate QEMU virtual machine running Alpine Linux on compromised Windows hosts. Because endpoint security agents cannot inspect inside the VM, attackers operate the full intrusion — credential theft, lateral movement, and data exfiltration — completely invisible to host-level detection.

#ransomware +8
💻 AppSec

Five-Year-Old ShowDoc RCE Flaw CVE-2025-0520 (CVSS 9.4) Now Under Active Exploitation — Over 2,000 Instances Exposed

Threat actors are actively exploiting CVE-2025-0520, a critical unauthenticated remote code execution vulnerability in ShowDoc — an IT documentation tool used by developers and operations teams. The flaw, patched in October 2020 but present in thousands of unupgraded installations, allows file upload exploitation to deploy web shells. More than 2,000 publicly accessible ShowDoc instances remain vulnerable.

#rce +7
🛡️ SecOps

April Patch Tuesday Bug Crashes LSASS on PAM-Enabled Domain Controllers — No Fix Yet

KB5082063, Microsoft's April 2026 cumulative update, is causing LSASS to crash on non-Global Catalog domain controllers in Privileged Access Management environments, triggering unrecoverable reboot loops that take down Active Directory authentication. Microsoft has confirmed the issue across all Windows Server versions from 2016 to 2025 and is developing a corrected update, but none is available yet.

#patch-tuesday +5
🔑 IAM

CISA Confirms Active Exploitation of Windows Task Host Privilege Escalation CVE-2025-60710 — Four Public Exploits Available

A link-following flaw in the Windows Host Process for Tasks allows any local user to escalate to SYSTEM privileges. Patched in November 2025, CVE-2025-60710 has been confirmed as actively exploited — CISA added it to the Known Exploited Vulnerabilities catalogue on 13 April with a 27 April federal deadline. Four public proof-of-concept exploits are now freely available on GitHub.

#privilege-escalation +5
🔬 Assessment

NIST Ends Full NVD Enrichment — What It Means for Your Vulnerability Management Programme

NIST has announced it will no longer enrich every CVE record in the National Vulnerability Database, shifting to a risk-based model that prioritises only the most critical submissions. With CVE volumes up 263% since 2020 and the NVD backlog now officially unresolvable, security teams that rely on NVD CVSS scores and CPE data for vulnerability prioritisation must urgently adapt their tooling and workflows.

#nvd +5
🗄️ Assets

Standard Bank Breach: 1.2TB of Client Data — Including Credit Card Details — Published Online

A threat actor claiming to have spent three weeks inside Standard Bank's network has published approximately 1.2TB of stolen data online, including client names, national identity numbers, account details, and a subset of credit card numbers. One of Africa's largest banks, Standard Bank operates across more than 20 countries and holds significant international exposure. The double-extortion attack pattern and lessons for database-layer monitoring are directly relevant to financial services defenders globally.

#data-breach +5