End-of-Day report
Timeframe: Mittwoch 25-03-2026 18:00 - Donnerstag 26-03-2026 18:30
Handler: Michael Schlagenhaufer
Co-Handler: Guenes Holler
News
Abusing Modern Browser Features for Phishing
Modern browsers have come a long way from rendering marked-up text to being used as a runtime for client applications. Many of the modern APIs require explicit user consent (e.g. when accessing Bluetooth, USB devices, location or local files) while others are allowed by default. As security researchers we are always interested in the worst-case - so what can malicious websites do without user consent? As it turns out - a very legitimate-looking phishing attempt! [..] We have reported this issue to both the Chromium project and Mozilla in May 2024. Though the issues have been accepted, no patch timeframe, solution approach or any other piece of information that would lead us to believe this issue will be resolved at any point was provided for almost two years. We have thus decided to publish the vulnerability details.
https://certitude.consulting/blog/en/abusing-modern-browser-features-for-phishing
TikTok for Business accounts targeted in new phishing campaign
Threat actors are targeting TikTok for Business accounts in a phishing campaign that prevents security bots from analyzing malicious pages.
https://www.bleepingcomputer.com/news/security/tiktok-for-business-accounts-targeted-in-new-phishing-campaign/
Claude Extension Flaw Enabled Zero-Click XSS Prompt Injection via Any Website
Cybersecurity researchers have disclosed a vulnerability in Anthropics Claude Google Chrome Extension that could have been exploited to trigger malicious prompts simply by visiting a web page. [..] Following responsible disclosure on December 27, 2025, Anthropic deployed a patch to the Chrome extension (version 1.0.41) that enforces a strict origin check requiring an exact match to the domain "claude[.]ai." Arkose Labs has since fixed the XSS flaw at its end as of February 19, 2026.
https://thehackernews.com/2026/03/claude-extension-flaw-enabled-zero.html
-Stack Overflow for AI Agents- Sounds Great - Until Someone Poisons the Answers
What if someone poisoned a doc in Context Hub-s registry? We simulated exactly that - built poisoned docs locally and served them through chub-s own MCP server. From the agent-s perspective, it is identical to the real thing. Haiku installed the fake dependency in 100% of runs. Warned the developer in 0%.
https://medium.com/@mickey.shmueli/stack-overflow-for-ai-agents-sounds-great-until-someone-poisons-the-answers-d322258095c4
Willhaben-Fake: Der Chatbot, der das Konto leerräumt
Eine SMS aus heiterem Himmel. Ein gefaktes Willhaben-Abrechnungsportal. Ein KI-Chatbot, der konkret auf Nachfragen eingeht. Die Kombination dieser drei Komponenten ist es, die einen aktuellen Betrugsversuch im Namen von Willhaben so gefährlich macht. Geht der Plan der Kriminellen auf, räumen Sie am Ende das Konto ihres Opfers komplett leer.
https://www.watchlist-internet.at/news/willhaben-fake-chatbot/
New PXA Stealer Malware Targets Banks, Uses Telegram to Exfiltrate Data
Financial firms across the globe are facing a fresh wave of digital break-ins this year. According to cyber threat detection firm CyberProof, a relatively new malware known as the PXA Stealer has seen a sudden spike in activity. During the first quarter of 2026, experts tracked an 8% to 10% increase in attacks using this specific tool.
https://hackread.com/financial-firms-rise-pxa-stealer-attacks/
Ghost Fleet: Half of All New Scanning IPs Last Week Geolocated to Hong Kong - Nearly None Completed a Connection
Last week, the GreyNoise Observation Grid observed something unusual: 242,666 new scanning IPs geolocating to Hong Kong appeared in seven days and 99.7% of them never completed a single TCP connection.
https://www.greynoise.io/blog/ghost-fleet-half-new-scanning-ips-geolocated-to-hong-kong
GhostClaw: Infostealer für macOS auf GitHub
Dieser versucht vom aktuellen Boom beim KI-Agenten OpenClaw zu profitieren und hofft, auf Nutzer zu treffen, die sich wenig mit dem Terminal auskennen - und dort Befehle einzutippen, die die Installation erst ermöglichen. Auch Entwickler, die nach OpenClaw-Werkzeugen suchen, könnten in die Falle tappen.
https://heise.de/-11222743
Vulnerabilities
Cisco Security Advisories 25.03.2026
Cisco has released 13 security advisories (1x critical, 5x high, 7x medium severity).
https://sec.cloudapps.cisco.com/security/center/publicationListing.x
Zwei kritische Schadcode-Lücken bedrohen Automatisierungsplattform n8n
Wie aus dem Sicherheitsbereich der GitHub-Website des Tools hervorgeht, haben die Entwickler insgesamt sechs Sicherheitslücken geschlossen. Zwei davon gelten als -kritisch- (CVE-2026-33696, CVE-2026-33660). Im ersten Fall kann nach einer Prototyp-Pollution-Attacke Schadcode auf Systeme gelangen und sie kompromittieren. Im zweiten Fall kann das ebenfalls möglich. Dieses Mal, weil AlaSQL sandbox bestimmte SQL-Anweisungen nicht ausreichend eingeschränkt.
https://www.heise.de/news/Zwei-kritische-Schadcode-Luecken-bedrohen-Automatisierungsplattform-n8n-11226060.html
Trend Micro Apex One- Service Pack 1 (2025) und kritischer Patch Build 17079 verfügbar
Trend Micro hat zum 26. März 2026 sowohl sein Trend Micro Apex One- Service Pack 1 (2025) veröffentlicht, als auch den kritischen Patch Build 17079 freigegeben. Letzterer enthält die gleichen Sicherheitsfixes wie das Service Pack 1 (2025), steht aber als Upgrade-Paket für bestehende Apex One-Installationen zur Verfügung. Sonderlich viele Informationen gibt es nicht.
https://borncity.com/blog/2026/03/26/trend-micro-apex-one-service-pack-1-2025-und-kritischer-patch-build-17079-verfuegbar/
TP-Link: Security Advisory on Multiple Vulnerabilities on TP-Link Archer NX200, NX210, NX500 and NX600 (CVE-2025-15517 to CVE-2025-15519 and CVE-2025-15605)
https://www.tp-link.com/us/support/faq/5027/
node.js: March 24, 2026 Security Releases
https://nodejs.org/en/blog/vulnerability/march-2026-security-releases
GitLab Patch Release: 18.10.1, 18.9.3, 18.8.7
https://about.gitlab.com/releases/2026/03/25/patch-release-gitlab-18-10-1-released/
LWN Security updates for Thursday
https://lwn.net/Articles/1064761/
Vienna Assistant: Local Privilege Escalation in Vienna Assistant (MacOS) - Vienna Symphonic Library
https://sec-consult.com/de/vulnerability-lab/advisory/local-privilege-escalation-in-vienna-assistant-macos-vienna-symphonic-library/