Hello, stay ahead with CISO Brief 🚀

⚡ This week's recap covers supply-chain compromises, resurfacing legacy bugs, and security tools themselves being targeted. Key incidents include a poisoned Nx Console VS Code extension leading to a GitHub breach, new active exploitation of Microsoft Defender flaws, and a nine-year-old Linux kernel privilege bug. Teams face increasing targeted phishing and widespread botnet scanning, while organizations scramble to patch critical CVEs and secure exposed services.

🛡️ Google Threat Intelligence Group analyzed a growing Chinese‑language phishing‑as‑a‑service (PhaaS) ecosystem, finding mature, professional offerings that facilitate real‑time credential and OTP interception and the tokenization of payment data. These services use encrypted channels like RCS and iMessage, provide extensive localization tools and ancillary criminal services, and often operate openly on Telegram. GTIG highlights the shift from simple password harvesting to financial account takeover and recommends stronger technical defenses such as FIDO2/WebAuthn and risk‑based verification.

🛡️ Dutch authorities arrested two co-owners of related hosting companies and seized over 800 servers on May 18, alleging they operated infrastructure used by Russia for cyberattacks and influence operations targeting the EU. The arrests follow investigative reporting that linked MIRhosting and WorkTitans to Stark Industries, an ISP sanctioned by the EU for facilitating DDoS, proxy, and anonymity services tied to Russia-backed actors. Officials searched businesses and data centers and charged the suspects with violating sanctions law by making economic resources available to sanctioned entities. Both suspects deny wrongdoing and one company says it has paused services to the implicated client pending internal review.

🔒 The FBI warns about the Kali365 phishing-as-a-service platform that abuses OAuth device code authentication to hijack Microsoft 365 and Microsoft Entra accounts. Distributed via Telegram since April 2026, Kali365 enables low-skilled attackers to bypass MFA by tricking victims into authorizing device codes, then capturing OAuth tokens to access mailboxes and cloud apps. Researchers observed campaigns using phishing emails, AI-generated lures, and real-time dashboards, while the FBI advises blocking device code flows and preserving forensic evidence.

🛡️ Researchers argue enterprises must stop treating AI agents as trusted components and instead secure them as untrusted systems. The paper, authored by teams from Google, UC San Diego, UW–Madison and others, distills five systems-security principles—least privilege, tamper resistance, complete mediation, secure information flow, and human risk—and maps eleven real-world agent attacks to these violations. They caution that stacking ML guardrails is insufficient and propose research directions for separating instructions from data, verifiable least-privilege policies, and information-flow controls.

🔍 CVE Lite CLI is an OWASP-backed, open-source scanner for JavaScript and TypeScript lockfiles that emphasizes local, early feedback for dependency vulnerabilities. The tool inspects npm, pnpm, and Yarn lockfiles using OSV data, distinguishes direct vs transitive issues, and recommends practical upgrade paths. It is designed as a lightweight developer tool complementing, not replacing, enterprise SCA platforms and intentionally keeps core vulnerability analysis deterministic while offering AI as an explanatory layer.

🛡️ The FBI has identified a new phishing-as-a-service platform called Kali365, first seen in April 2026, that is being distributed primarily via Telegram. The service furnishes AI-generated lures, automated templates and real-time tracking dashboards to enable attackers — including low-skill actors — to capture OAuth tokens and bypass MFA for Microsoft 365 accounts. Victims are tricked into pasting device codes into the legitimate Microsoft verification page, unintentionally authorizing attacker devices and granting persistent access to services such as Outlook, Teams and OneDrive. The FBI recommends restricting or blocking device code flow, implementing conditional access policies, blocking authentication transfer and protecting emergency access accounts.

🚨 A Bitdefender report warns that cybercriminals have built extensive ecosystems to scam Formula 1 fans, exploiting the sport’s fast-moving digital culture. Scams include counterfeit merchandise, fake grand prix tickets, illegal streaming apps and boxes, social media fraud and distribution of infostealer malware. Fans may also be coerced into botnets for DDoS attacks. Bitdefender urges vigilance and recommends anti-phishing and antivirus tools to reduce risk.

🔒 A survey of 750 CISOs in the US and UK found 58% said their organization would be willing to pay a ransom to end a ransomware incident. Experts and law enforcement advise against paying, citing encouragement of attackers and no guarantee of data recovery, but real-world evidence shows many firms still pay. Industry sources report incomplete decryption and credential exposure even after payment, while robust backups remain the best mitigation.

🛡️ Researchers uncovered a large-scale campaign exploiting a critical SQL injection (CVE-2026-26980) in Ghost CMS to inject malicious JavaScript that triggers ClickFix attack flows. More than 700 domains — including university portals, media outlets, fintech firms, and personal blogs — were affected. The flaw impacts Ghost 3.24.0 through 6.19.0 and allows unauthenticated actors to exfiltrate admin API keys. Administrators are urged to upgrade to 6.19.1+, rotate keys, and scan sites for injected scripts.

🔒 GitHub has introduced staged publishing on npm, requiring a human maintainer to complete a two-factor authentication (2FA) challenge before a package version becomes publicly installable. The prebuilt tarball is uploaded to a staging queue and only becomes available after explicit approval. Maintainers must have publish access, an existing package, and enabled 2FA. GitHub also added three install-source flags to control non-registry installs.

🔎 Anthropic disclosed that Project Glasswing and access to Claude Mythos Preview helped partners uncover over 10,000 high- or critical-severity vulnerability candidates across widely used, systemically important software since last month. Analysis verified 1,726 true positives, including 1,094 high- or critical-severity flaws, and resulted in 97 upstream patches and 88 advisories. One notable finding was a critical WolfSSL flaw (CVE-2026-5194).

🔐 A critical vulnerability, CVE-2026-48172 (CVSS 10.0), in the LiteSpeed User-End cPanel Plugin allows privilege escalation via the lsws.redisAble function, enabling arbitrary scripts to run as root. The flaw affects plugin versions 2.3 through 2.4.4 and is being actively exploited; LiteSpeed fixed it in v2.4.5 and later bundled releases. Administrators are urged to upgrade to cPanel plugin v2.4.7 (with WHM plugin v5.3.1.0) or uninstall the user-end plugin if immediate patching is not feasible.

🛡️ The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has added a critical SQL injection flaw in Drupal Core (CVE-2026-9082, CVSS 6.5) to its Known Exploited Vulnerabilities list after evidence of active exploitation. The vulnerability affects all supported Drupal Core versions and could enable privilege escalation and remote code execution via crafted requests using the database abstraction API. Patches were released across multiple 8.x–11.x branches, with manual patches required for Drupal 9.5 and 8.9.

🛡️ Chromium contains an unpatched vulnerability that lets attackers keep a Service Worker alive across restarts and execute JavaScript persistently. Reported by researcher Lyra Rebane, the bug abuses the Background Fetch API and a race that creates and aborts background fetches to evade UI visibility. Although some UI fixes were applied in 2023, the deeper issue—preventing indefinite Service Worker lifetimes—remains unresolved and can enable tracking, crypto mining, and browser-based bots.

🔒 Amazon SageMaker Unified Studio now supports domain management for both Identity Center and IAM-based domains outside the AWS Console. Administrators and data management teams can create and manage projects, configure workforce identity, administer users and permissions, and set networking properties. VPC configuration and account associations are consistent across domain types and available in all Regions where Unified Studio is offered.

🔍 AWS Transform now includes enhanced migration assessment capabilities that support what-if scenarios, customizable assumptions, flexible file formats, and expanded TCO assessment features. These updates enable rapid building of migration business cases and faster decision-making. The tool accepts inputs from RVTools, CMDBs, AWS discovery exports, and many third-party discovery tools. New analysis options cover EC2, FSx, S3, SQL Server on EC2, virtual desktops, and additional Cloud Value Framework pillars.

🛠️ Amazon SageMaker Unified Studio now adds business context, metadata, and data governance features for IAM-based domains. Customers can annotate AWS Glue Data Catalog tables with business names, descriptions, and README documentation, and use AI-generated metadata to automate cataloging. Teams can build business glossaries, define metadata form templates, and capture structured attributes like classification, retention, and ownership. These capabilities enable search, filtering by glossary or metadata fields, and access requests with automated Lake Formation permission grants, and are available in all regions where SageMaker Unified Studio is supported.