Vulnerabilities
11 stories[NEU] [kritisch] Citrix NetScaler ADC und Citrix NetScaler Gateway: Mehrere Schwachstellen
BSI rates new Citrix NetScaler ADC/Gateway flaws as critical — enabling RCE, data manipulation, and security bypass on a device class repeatedly targeted by state-sponsored actors.
CERT-Bund (BSI)9/10[NEU] [hoch] IBM QRadar SIEM: Mehrere Schwachstellen
Multiple high-severity IBM QRadar SIEM vulnerabilities allow code execution and privilege escalation—compromising the very tool SOCs depend on for threat detection and log integrity.
CERT-Bund (BSI)9/10Cisco warns of new SD-WAN zero-day exploited in attacks
Cisco disclosed CVE-2026-76504, an actively exploited zero-day in Catalyst SD-WAN Manager that lets attackers bypass authentication and escalate to admin—demanding emergency patching across enterprise WAN estates.
BleepingComputerCVE-2026-765049.89/10Fortinet warns of critical FortiMail flaw exploited in zero-day attacks
Fortinet confirms active zero-day exploitation of a critical FortiMail RCE (CVE-2026-104286)—European enterprises running FortiMail should treat this as an emergency given the vendor's pattern of perimeter-appliance compromises.
BleepingComputerCVE-2026-1042869.89/10Threat Brief: NetScaler Zero Days CVE-2026-88771 and CVE-2026-88772 Exploited in the Wild
Two Citrix NetScaler zero-days (CVE-2026-88771, CVE-2026-88772) are being exploited in the wild, giving attackers unauthenticated access to perimeter infrastructure—patch or isolate affected appliances without delay.
Unit 42 (Palo Alto)CVE-2026-887719.89/10Attackers Exploit Zimbra Flaw to Deploy Web Shells and Harvest Authentication Secrets
CVE-2026-73570 (CVSS 8.9) in Zimbra is under active exploitation—attackers deploy web shells and harvest authentication secrets from unpatched instances still common in European organisations.
THN (Feedburner)CVE-2026-735708.99/10GitLab warns of critical RCE vulnerability in AI Gateway service
GitLab's critical RCE in its AI Gateway service puts every connected CI/CD pipeline at supply-chain risk—patch immediately or disable the feature until remediation is confirmed.
BleepingComputer9/10[NEU] [hoch] JetBrains TeamCity: Mehrere Schwachstellen
CERT-Bund issues a high-severity alert for JetBrains TeamCity with RCE and admin-escalation flaws — a direct threat to CI/CD supply-chain integrity for any organisation running on-prem instances.
CERT-Bund (BSI)9/10[NEU] [UNGEPATCHT] [kritisch] Foreman: Mehrere Schwachstellen
BSI warns of critical, still-unpatched Foreman vulnerabilities allowing RCE and security-bypass—a serious risk for enterprises using it to automate server provisioning at scale.
CERT-Bund (BSI)8/10[NEU] [hoch] Tenable Security Nessus: Mehrere Schwachstellen
Multiple high-severity Tenable Nessus flaws—including RCE, SQLi, and privilege escalation—risk turning your own vulnerability scanner into an attack vector if left unpatched.
CERT-Bund (BSI)8/10Kiteworks & Citrix Incidents Show Challenges of Zero-Day Response
Kiteworks forced a nine-hour platform shutdown while Citrix stayed silent on active attacks before patching—both cases expose how vendor zero-day communication failures compound enterprise risk.
Dark Reading8/10
Threat Intel
5 storiesFrench Tax Data Theft Using Stolen Staff Passwords Went Undetected for Seven Weeks
Attackers used stolen staff credentials to siphon tax data on hundreds of thousands of French citizens for seven weeks before detection—exposing critical gaps in credential governance and exfiltration monitoring at a major EU public body.
THN (Feedburner)9/10Bring Your Own Vulnerable Device: The EDR Killer Economy
SentinelOne maps how BYOVD tactics have matured from ad-hoc driver abuse into a tiered commercial service economy selling EDR-killer tools—raising the baseline threat to every endpoint-reliant defence.
SentinelOne Blog9/10Russia's Star Blizzard Ditches ClickFix to Widen Phishing Net
Russia's Star Blizzard APT has abandoned ClickFix in favour of a new phishing method called RedFlick, deploying a CosmicPulse backdoor against Ukrainian-linked NGOs, think tanks, and journalists—broadening its target aperture.
Dark Reading8/10China-Aligned TA419 Targets U.S. AI Policy Experts With Microsoft AitM Phishing
China-nexus group TA419 runs adversary-in-the-middle phishing against US AI policy researchers, bypassing MFA—a technique likely to extend to European regulatory stakeholders as the EU AI Act takes effect.
THN (Feedburner)8/10Malicious Linux Implants Mimic Asian Mail Security Products
Three newly identified Linux backdoors impersonate legitimate Asian mail-security edge products, making network-level detection extremely difficult—edge-appliance integrity verification is now essential.
Dark Reading8/10
AI Security
5 storiesCheap to Hypothesize, Costly to Verify: The Defense Surface of Agentic Vulnerability Discovery
This paper formalises the economic asymmetry of LLM-driven vulnerability hunting — hypotheses are cheap but verification is costly — giving security leaders a framework to evaluate agentic scanning ROI realistically.
arXiv Crypto & Security9/10Approval Laundering: Systematizing Approval--Execution Binding Failures in AI Coding-Agent Harnesses
Major AI coding agents—Claude Code, Codex CLI, Cursor—suffer from approval-laundering flaws where the action a developer approves silently diverges from what executes, undermining human-in-the-loop security assumptions.
arXiv Crypto & Security9/10Autonomous OSS Threat Detection via Taxonomy-Aligned LLMs
Researchers use taxonomy-aligned LLMs to automatically detect OSS supply-chain threats like typosquatting and CI/CD poisoning—an approach that could augment SCA tools where manual review doesn't scale.
arXiv Crypto & Security8/10Malicious Email Could Hijack AI Agent and Access Connected Accounts
A patched flaw in the Manus agentic-AI platform allowed a crafted email to hijack the agent and pivot into connected accounts—one of the first documented prompt-injection-to-lateral-movement chains in production.
IT Security Guru8/10Tokenized Key-Gated Adapter Routing: A Secure Access Control Mechanism Against Private Data Leakage in LLMs
Researchers introduce tokenized key-gated adapter routing to prevent LLMs from leaking memorised PII—a pattern with direct implications for GDPR-compliant AI deployments in healthcare, finance, and government.
arXiv Crypto & Security7/10
Regulatory
2 storiesAn Empirical Study and Assessment of EU AI Act Compliance Checkers
An empirical assessment of existing EU AI Act compliance-checking tools finds significant coverage gaps in security, privacy, and risk-management requirements — essential reading as enforcement timelines approach.
arXiv Crypto & Security9/10New Mexico jury finds Meta deceived consumers about data privacy practices
A New Mexico jury found Meta liable for roughly 44 million privacy-law violations tied to deceptive data practices—a precedent that could embolden EU regulators pursuing similar GDPR enforcement against Big Tech.
The Record8/10
Research
1 storyCloud
1 storySecurity
1 storyCompliance
1 storyDevSecOps
1 storyBoardroom Brief
What this week's reporting means for the board, in one line per story.
- French Tax Data Theft Using Stolen Staff Passwords Went Undetected for Seven Weeks
A simple stolen-password attack on France's tax authority went unnoticed for seven weeks, reinforcing that credential governance is a board-level priority.
- An Empirical Study and Assessment of EU AI Act Compliance Checkers
Current EU AI Act compliance-checking tools have significant blind spots — relying on them without validation creates regulatory exposure.
- [NEU] [kritisch] Citrix NetScaler ADC und Citrix NetScaler Gateway: Mehrere Schwachstellen
Critical vulnerabilities in Citrix NetScaler — a key perimeter component — require emergency patching to prevent potential network breach.
- [NEU] [hoch] IBM QRadar SIEM: Mehrere Schwachstellen
Vulnerabilities in IBM's QRadar SIEM could allow attackers to compromise the organisation's central security monitoring capability.
- Cheap to Hypothesize, Costly to Verify: The Defense Surface of Agentic Vulnerability Discovery
AI-driven vulnerability scanning produces many leads but verifying them is expensive — budget and process guardrails are essential.
- Cisco warns of new SD-WAN zero-day exploited in attacks
An actively exploited Cisco SD-WAN zero-day allowing full admin takeover requires immediate patching to protect enterprise network infrastructure.
- Fortinet warns of critical FortiMail flaw exploited in zero-day attacks
An actively exploited zero-day in Fortinet's FortiMail product allows remote code execution—boards should confirm emergency patching is underway and assess exposure.
- Threat Brief: NetScaler Zero Days CVE-2026-88771 and CVE-2026-88772 Exploited in the Wild
Two actively exploited zero-day vulnerabilities in Citrix NetScaler require immediate patching to prevent perimeter breach.
- Attackers Exploit Zimbra Flaw to Deploy Web Shells and Harvest Authentication Secrets
Attackers are actively exploiting a critical Zimbra vulnerability to steal email credentials—immediate patching and breach assessment are required.
- GitLab warns of critical RCE vulnerability in AI Gateway service
A critical vulnerability in GitLab's AI service could let attackers inject code into CI/CD pipelines—immediate patching is required to protect the software supply chain.
- [NEU] [hoch] JetBrains TeamCity: Mehrere Schwachstellen
High-severity vulnerabilities in JetBrains TeamCity could let attackers execute code and gain admin rights on CI/CD systems — immediate patching is critical to protect the software supply chain.
- Approval Laundering: Systematizing Approval--Execution Binding Failures in AI Coding-Agent Harnesses
AI coding agents used by development teams can silently execute actions that differ from what was approved, creating an integrity gap in software delivery.
- OperTraitors: How Kubernetes Operators Betray Your Security Posture
Over-privileged Kubernetes automation accounts represent a material lateral-movement risk in containerised infrastructure.
- Bring Your Own Vulnerable Device: The EDR Killer Economy
Commercially available tools that disable endpoint security are proliferating; board-level investment in defence-in-depth beyond EDR is warranted.
- [NEU] [UNGEPATCHT] [kritisch] Foreman: Mehrere Schwachstellen
Critical unpatched vulnerabilities in Foreman provisioning software could allow attackers to execute code across managed infrastructure.
- Russia's Star Blizzard Ditches ClickFix to Widen Phishing Net
Russia-linked Star Blizzard has evolved its phishing tradecraft, expanding targeting to civil-society groups relevant to European policy circles.
- China-Aligned TA419 Targets U.S. AI Policy Experts With Microsoft AitM Phishing
A Chinese espionage group is actively stealing credentials from AI policy experts—EU regulatory personnel are a logical next target as the AI Act rolls out.
- [NEU] [hoch] Tenable Security Nessus: Mehrere Schwachstellen
Flaws in the Nessus vulnerability scanner could let attackers leverage our own security tooling for reconnaissance and code execution.
- The Fine Art of Frustrating the Adversary
Cisco Talos researchers outline practical ways to slow attackers and raise their operational costs — a low-investment uplift to existing defences.
- Kiteworks & Citrix Incidents Show Challenges of Zero-Day Response
Recent Kiteworks and Citrix zero-day incidents demonstrate that vendor communication failures during active exploitation can force unplanned outages and extend exposure windows.
- Autonomous OSS Threat Detection via Taxonomy-Aligned LLMs
Automated detection of open-source supply-chain threats is becoming essential as attack sophistication outpaces manual code review capacity.
- Agents Pick Dependencies, DoWI 8430.01 Holds You Accountable
Regulators now hold organisations liable for software dependencies chosen by AI agents — review your automated development governance.
- Malicious Email Could Hijack AI Agent and Access Connected Accounts
A newly disclosed attack chain shows that a single malicious email can commandeer an AI agent and access linked enterprise accounts, requiring immediate review of AI-agent permissions.
- Ship fast, verify independently: keeping application security in step with AI-written code
AI-generated code is shipping faster than security teams can verify it—independent testing must become a mandatory pipeline gate to avoid compliance and exploit risk.
- New Mexico jury finds Meta deceived consumers about data privacy practices
A US jury found Meta committed 44 million privacy violations—EU regulators may follow suit, making transparent data practices a board priority.
- Tokenized Key-Gated Adapter Routing: A Secure Access Control Mechanism Against Private Data Leakage in LLMs
A new technical control could help prevent LLMs from leaking personal data, addressing a growing regulatory and reputational risk.
Forward-looking interpretation drafted by editorial AI under human review — not a reproduction of the source. See methodology.