RovoBlast: How One Click Triggered Atlassian’s AI Assistant to Leak Data

Varonis Threat Labs disclosed RovoBlast, a one-click attack against Atlassian Rovo that injects attacker-controlled instructions through a crafted link and can make the assistant treat external parameters as trusted session input. The flaw can expose data across Jira, Confluence, Bitbucket, Slack, Microsoft 365, Google Workspace, and other connected systems, and Atlassian…

Read More
Unlimited Technology Systems breach impacts 3.8 million people

Unlimited Technology Systems disclosed that a breach of its commercial data center exposed personal and health-related information belonging to 3,803,750 people. The company said unauthorized access occurred for five days in October 2025, and it has not identified the attackers or any public ransomware group claiming the incident. #UnlimitedTechnologySystems

Read More
Metabase SQLi zero-day exploited in customer data-theft attacks

A critical unauthenticated SQL injection flaw in Metabase was actively exploited as a zero-day, giving attackers administrator access to customer instances and enabling data theft from both Metabase Cloud and self-hosted deployments. Framework, Tally, and LexisNexis all reported impacts tied to the incident, with stolen data including customer contact details, login information, and in some cases password hashes. #Metabase #Framework #Tally #LexisNexis

Read More
UNC6671 Vishing Attacks Target Personal Phones to Steal SaaS Data

UNC6671 is carrying out vishing-led extortion campaigns against financial services, private equity, and professional services by impersonating IT help desk staff and stealing credentials through fake login portals. The group uses captured sessions to access Microsoft 365 and Okta, exfiltrate cloud data, and operate under multiple extortion brands including Redact, Pink,…

Read More
ClickFix Attacks Deliver macOS Stealer That Can Drain Crypto Wallets

ClickFix-style lures are being used to deliver a Go-based macOS stealer that can harvest browser passwords, Apple iCloud Keychain data, cached credentials, and cryptocurrency holdings. The campaign also uses Aeza Group infrastructure and joins a broader wave of ClickFix attacks tied to MacSync, Atomic Stealer, Lumma Stealer, and Remus. #ClickFix #AezaGroup…

Read More
APT37 Strikes Again, This Time with NarwhalRAT

Genians Security Center researchers uncovered a new APT37 campaign using NarwhalRAT to steal data through keylogging, screen capturing, USB data collection, and remote code execution. The intrusion likely began with spearphishing emails and malicious LNK files, while the operators used a Korean relay server and the pCloud API in a dual C&C structure to support the attack. #APT37 #NarwhalRAT #pCloud

Read More
More than half of AI-generated patches are broken

New research from 1Password found that ChatGPT 5.5 and Claude Opus 4.8 often fail to fully patch high-impact vulnerabilities and can introduce new bugs instead. The findings suggest AI-assisted patching still needs human review, even as models like GPT 5.5, GPT-5.6-Sol, and Anthropic’s Mythos are promoted for stronger security capabilities. #ChatGPT #ClaudeOpus #1Password #Veracode #ProjectGlasswing #Daybreak

Read More
Ransom! Hitech Distribuzione Informatica S.r.l. (HTDI) (AUG-2026)

spacebears is reported to have targeted Hitech Distribuzione Informatica S.r.l. (HTDI), a leading Italian IT solutions provider based in Rome (Via Tempio del Cielo) offering turnkey infrastructure services across design, delivery, installation, integration, and ongoing operations for corporate clients. The ransomware claim implicates HTDI as the victim in Italy’s IT services sector. #Italy

Read More
Beyond ‘Fake Updates’: From Application Store-Themed Phishing to Large-Scale Distribution of ScreenConnect

LevelBlue OpsCTI identified a large-scale phishing campaign that impersonates trusted apps and services to trick victims into installing unauthorized ConnectWise ScreenConnect clients through fake update and installation flows. The campaign uses compromised sites, attacker-controlled hosts, Amazon S3, Cloudflare R2, Telegram telemetry, and reused infrastructure artifacts to deliver remote access payloads and expand hunting opportunities. #ConnectWiseScreenConnect #GoogleMeet #MicrosoftStore #AppleAppStore #TelegramBotAPI

Read More
New WordPress Pre-Auth XSS Could Lead to PHP Code Execution – Patch ASAP

WordPress has patched CVE-2026-64638, a pre-authentication reflected XSS flaw in the login screen that affects every version of the CMS and can be chained into PHP code execution under the right conditions. Researchers at pwn.ai demonstrated the attack chain XSS2Shell, but WordPress says real-world escalation still depends on administrator interaction and…

Read More
In Other News: AI Slop Limits Apple Bounties, North Carolina Port Attacks, Hackers Target Wall Street

SecurityWeek’s weekly roundup covers major cybersecurity developments, including OpenAI’s disruption of a Cambodia-based scam network, Amgen’s cloud data theft incident, and Apple’s move to limit AI-generated bug bounty noise. It also highlights supply chain compromises, router backdoors, phishing and vishing attacks, and policy efforts targeting Chinese data center components. #OpenAI #Amgen…

Read More