China-Linked Hackers Targeted University Email Servers

A threat cluster tracked as UNK_MassTraction, believed to be aligned with Chinese state interests, has been running a targeted campaign against physics and engineering departments at U.S. and Canadian universities, according to The Hacker News. The campaign was first detected in May 2026 and singled out administrators and professors with ties to national security research, astrophysics, and particle physics.

The attackers exploited two critical flaws in Roundcube, an open-source webmail platform commonly used by universities. Both CVE-2024-42009 and CVE-2025-49113 had already been patched, but the targeted departments were still running vulnerable versions — suggesting the attackers did their homework first and identified targets before sending a single email.

The attack required no unusual action from victims. Opening the malicious email inside Roundcube was enough. A payload called IceCube executed silently in the browser, harvesting saved credentials, two-factor authentication (2FA) codes, and session cookies. Think of it like a thief who only needs you to open your front door — not even step outside.

From there, IceCube used the stolen session data to trigger a second, more severe remote code execution (RCE) flaw, allowing the attackers to run code directly on the mail server. The primary tool deployed was VShell, a known post-exploitation tool previously linked to a China-nexus cluster called UNC5174. A fallback web shell called SquareShell was also used when VShell deployment failed. An ELF loader called SNOWLIGHT helped deliver the tools and has appeared in other Chinese-attributed intrusions.

What you should do: If your organisation runs Roundcube, confirm it is updated to the latest version immediately. Any university IT team should audit mail server logs for unexpected outbound HTTP POST requests or access to unusual plugin endpoints.

Researchers at Zscaler's ThreatLabz have documented two real-world campaigns using a technique called indirect prompt injection, where attackers hide instructions inside web pages that AI agents read and trust, steering the agent's behaviour without the user ever knowing, according to Infosecurity Magazine.

In both cases, attackers used SEO poisoning (the manipulation of search rankings to push malicious sites to the top of results) to ensure their pages were found. Instructions were buried using CSS to move text off-screen or tucked into structured metadata — invisible to humans, readable by machines.

One fake page posed as Python software documentation and instructed AI coding agents to purchase a bogus API key via cryptocurrency. Four of 26 large language models (LLMs) tested were successfully manipulated into completing the fraudulent payment, including versions of Meta's Llama and Google's Gemini.

What you should do: If you use AI agents for coding or research tasks, treat any payment instruction surfaced by an agent as a red flag requiring manual verification.

What Adobe ColdFusion is: A commercial platform used by developers to build and deploy enterprise websites and web applications.

What it is: A maximum-severity flaw allowing an unauthenticated attacker — one with no account or prior access — to execute arbitrary code remotely on an unpatched ColdFusion server. Adobe released a patch and urged administrators to apply it within 72 hours, per Bleeping Computer.

Who's at risk: Any organisation running ColdFusion versions 2025.9, 2023.20, or earlier with internet-facing deployments.

CVSS: 10.0 (Critical — patch today, full stop.)

Root cause: The flaw requires no authentication and no user interaction to trigger. An attacker simply sends a crafted request to the server. The underlying failure is in how ColdFusion handles external input before verifying who is sending it, allowing code execution before any identity check occurs.

Attack vector: Exploitation began within two hours of Adobe's public disclosure, captured by KEVIntel's global honeypot network. An attacker sends a malicious HTTP request to an exposed ColdFusion instance, gains code execution, and can subsequently read files, install backdoors, or pivot deeper into the network. No credentials needed.

Detection strategies: Look for unexpected outbound connections from your ColdFusion server, unusual process spawning from the ColdFusion service account, and unfamiliar files written to the web root directory.

Recommended actions:

1. Apply Adobe's patch immediately — the 72-hour window has already passed for most organisations.
2. Place ColdFusion servers behind a web application firewall (WAF) if not already done.
3. Restrict external access to ColdFusion admin interfaces at the network level.

Microsoft is enabling its Windows settings backup tool by default for organisations running Windows 11 26H2, according to Bleeping Computer. The feature, available on Microsoft Entra-joined devices, automatically backs up a user's Windows settings so they can be restored after a device is reset, replaced, or wiped.

This matters from a security standpoint: when employees lose a device or face a ransomware recovery, having settings backed up cuts recovery time significantly and reduces the window of exposure.

What you should do: If your organisation uses Microsoft Entra (formerly Azure Active Directory), check whether the backup policy is configured intentionally. IT administrators can manage the setting through Microsoft Intune or Group Policy. If your organisation operates in an EU Digital Markets Act region, the default-on behaviour does not apply — but enabling it manually is still worth considering.

The UK government has launched a Cyber Resilience Pledge, with over 60 organisations including Marks & Spencer, Nationwide, and Microsoft UK committing to board-level cybersecurity accountability, per Infosecurity Magazine. Signatories must register for the NCSC's free Early Warning alert service and push Cyber Essentials certification (a government-backed security baseline standard) down their supply chains.

For smaller suppliers, this creates real pressure to meet a minimum security standard to retain contracts with larger partners. Businesses with under £20m turnover that achieve Cyber Essentials certification are eligible for free cyber-liability insurance.