Why Should I Care? β 2026-10-27 | π΄ 2 HIGH Β· π‘ 0 MEDIUM Β· π΅ 16 RADAR Β· βͺ 187 FILTERED
π Briefing β 2026-10-27
18 vendor intel items scanned | π΄ 2 HIGH | π‘ 0 MEDIUM | π΅ 16 RADAR | βͺ 187 FILTERED
π΄ Critical β action required:
- CISA Adds Two Known Exploited Vulnerabilities to Catalog (CVE-2026-109618, CVE-2026-109619) β Yes, if you run PHP: These vulnerabilities are actively exploited in the wild and allow for remote code execution or privilege escalation, requiring immediate patching.
- Multiple Node.js CVEs (CVE-2026-106069) β Yes, if you run FortiOS: Fortinet is actively investigating whether the underlying Node.js vulnerabilities affect your appliances, and while many products are confirmed safe, FortiOS remains a potential risk until patched.
Everything else can wait.
π΅ 15 items on the radar β see below β
π΄ HIGH β Handle Now
CISA Adds Two Known Exploited Vulnerabilities to Catalog
CISA Advisories [CISA KEV] | CVE-2026-109618, CVE-2026-109619
β Why Should I Care?
Yes, if you run PHP: These vulnerabilities are actively exploited in the wild and allow for remote code execution or privilege escalation, requiring immediate patching.
π― Affected versions: PHP versions prior to the latest stable release (specific version ranges not detailed in this advisory, but implied to be all unpatched versions)
Not affected: PHP versions containing the latest security patches
π In plain English:
These bugs allow attackers to break the internal logic of your web server, potentially letting them run malicious code or access data they shouldn't. For example, an attacker could send a specially crafted request to your website to take full control of the server.
π§ Prerequisites:
- Running a vulnerable version of PHP
- Web server is exposed to the internet or internal network
- Attacker can send HTTP requests to the PHP application
β± Urgency: Critical: These vulnerabilities are in the CISA KEV catalog, meaning they are being actively exploited by threat actors right now.
β Fixed in: Latest stable PHP release
π‘ Context: The root cause involves memory corruption and stale internal request-state handling in the PHP engine, which can be triggered by malicious input.
Multiple Node.js CVEs
Fortinet PSIRT | CVSS 9.8 | CVE-2026-106069
β Why Should I Care?
Yes, if you run FortiOS: Fortinet is actively investigating whether the underlying Node.js vulnerabilities affect your appliances, and while many products are confirmed safe, FortiOS remains a potential risk until patched.
π― Affected versions: FortiOS (under investigation); Node.js 22.x, 24.x, 26.x
Not affected: FortiManager, FortiAnalyzer, FortiClient EMS, FortiAuthenticator, FortiWeb, FortiSOAR, FortiPresence, FortiADC, FortiCamera, FortiClient (iOS/MacOS/Windows), FortiDeceptor, FortiEDR (Android/iOS), FortiToken (Android/iOS/Windows), FortiVoice UC, FortiSIEM Windows Agent
π In plain English:
The Node.js runtime used by some Fortinet products has a critical flaw that allows an attacker to execute arbitrary code on the server. For example, an attacker could send a specially crafted request to a FortiOS management interface to gain full control of the firewall, potentially bypassing security controls or exfiltrating data.
π§ Prerequisites:
- Running a FortiOS version that has not yet been confirmed as unaffected or patched
- Attacker has network access to the vulnerable service
- No authentication is required for the attack (Unauthenticated)
β± Urgency: Monitor Fortinet PSIRT updates closely and prepare to patch immediately, as the underlying vulnerability allows unauthenticated remote code execution with a critical CVSS score of 9.8.
π‘ Context: The root cause is a flaw in the Node.js Brotli decoder that permits an out-of-bounds write, enabling remote code execution.
π‘ MEDIUM (0)
None.
π΅ On the Radar (16)
- University of Copenhagen data breach exposes data of 525,000 people (BleepingComputer) β Hackers accessed a test environment at the University of Copenhagen, exposing the names, national ID numbers, phone numbers, and emails of 525,000 people, plus PIN codes for 20,000 staff and students. This creates a high risk of targeted phishing and social engineering attacks against these individuals. For infrastructure professionals, the key takeaway is that 'test' environments are not safe havens if they contain real personal data.
- PHP Issues Second Windows Security Update in Five Days to Fix Brotli Code Execution Flaw (The Hacker News) β PHP's official Windows builds for versions 8.2 through 8.5 shipped with a vulnerable version of the Brotli compression library (1.2.0). If your application uses cURL to fetch data and automatically decompresses responses (which is common in frameworks and libraries like Guzzle), an attacker can send a maliciously crafted compressed response that corrupts memory and executes arbitrary code. This is a critical vulnerability (CVSS 9.2) that allows remote code execution without authentication.
- Hackers Infect 953 Online Stores With Scanner-Evading JavaScript, Mostly on Magento (The Hacker News) β Hackers have infected 953 online stores (mostly Magento) with malicious JavaScript hidden in CMS blocks. This code uses browser fingerprinting to detect security scanners and bots, serving them harmless content while delivering malicious payloads to real shoppers. The malware persists in the browser's localStorage, meaning it can re-execute even if the attacker's domain goes down. For example, a customer visiting your site might unknowingly have their session hijacked or be redirected to a phishing site, while your WAF sees nothing suspicious.
- Critical 'HTTP2Shell' Brotli vulnerability triggers emergency patches (BleepingComputer) β A buffer overflow in the Brotli compression library (CVE-2026-111497) allows attackers to corrupt memory and execute arbitrary code by serving maliciously compressed HTTP responses. For example, a Node.js API server fetching a URL from an untrusted source could be compromised if it decompresses a crafted Brotli stream, leading to full system takeover.
- DreamHost forces password resets in ongoing "incident" (BleepingComputer) β DreamHost is forcing password resets for email and file transfer services after an incident that disrupted their File Manager. This suggests attackers may have accessed accounts or added unauthorized email forwarding rules. If you rely on DreamHost for business communications or file transfers, you are currently locked out until you reset your password, and you must check for any suspicious forwarding rules that could be leaking your emails to attackers.
- Gheist Uses Stolen GitHub Tokens to Steal Private Repository Secrets via Actions Logs (The Hacker News) β Attackers are using stolen GitHub tokens to push a malicious workflow file (sync-check.yml) to a new branch (ci-sync-XXXXXX). This workflow prints all repository secrets in base32 encoding to the logs, which bypasses GitHub's secret masking. The attackers then download the logs and delete the run. This affects private repos and GitHub Enterprise, where GitHub's new 'pause suspicious runs' safeguard does not apply. If you have write access to a repo, your secrets are exposed.
- Former pentester used ARTEX AI to steal 8.2 million data records (BleepingComputer) β A former pentester used the ARTEX AI framework to breach over 90 organizations (including hospitals and corporations) and steal 8.2 million records. The key takeaway is that AI tools are enabling non-elite actors to execute complex, large-scale attacks that previously required specialized skills. For example, a single actor could now automate the discovery and exploitation of vulnerabilities across dozens of targets simultaneously, leading to massive data exfiltration and extortion demands.
- Attackers Use Microsoft Teams IT Support Scams to Deploy Malware and Ransomware (The Hacker News) β Attackers are creating fake Microsoft 365 tenants with names like 'IT Support' or 'Help Desk' to send Teams messages to your employees. They often flood the victim's inbox with spam first to create urgency, then ask for remote access via tools like AnyDesk or Quick Assist. Once inside, they install backdoors, steal data, and deploy Chaos ransomware. This bypasses traditional email security because the message comes from a valid (but fake) Microsoft domain.
- Malicious File in Downloads Folder Tries to Trick AI Coding Agent Into Installing Malware (The Hacker News) β Attackers are placing fake instruction files (like AGENTS.md) in folders that AI coding agents trust. When the agent starts in that folder, it reads the file as commands and may install malicious packages or redirect API traffic to steal your login tokens. For example, a developer opening Codex in their Downloads folder could unknowingly execute a malware installer because a web page silently dropped a malicious instruction file there.
- OpenAI launches GPT-6.1 Terra with 3x faster AI agents, 50% lower pricing (BleepingComputer) β New AI model launch with improved performance and lower pricing is relevant for cost optimization and capability upgrades.
- Two Critical Wasmtime Flaws Could Let Malicious WebAssembly Modules Escape Sandboxes (The Hacker News) β News report on critical sandbox escape flaws in Wasmtime.
- β‘ Weekly Recap: This2Shell, That2Shell, Everything's an RCE? Rogue AI, Supply Chain Chaos and More (The Hacker News) β Weekly summary of multiple critical RCE vulnerabilities and supply chain issues.
- Chinese man extradited for role in $36 million cryptocurrency scam (BleepingComputer) β Legal news regarding a cryptocurrency scam extradition.
- When an AI Agent Is the Incident, the Identity Record Is the Evidence (BleepingComputer) β Guidance article on incident response for AI agents.
- Microsoft releases emergency .NET patches for critical Brotli flaw (BleepingComputer) β News report on Microsoft's emergency patch for a critical .NET vulnerability.
βͺ 187 low-priority items filtered.
π¦ Aggregated and triaged by Donna AI | Sources: 8 vendor feeds | CISA KEV