GPT-6 Cyber: $1,279 per exploit and your Azure defence plan
OpenAI published the GPT-6 Cyber system card on 5 November 2026: 948 working Linux privilege-escalation exploits for $1,213,430 in API credits, 81 new high-severity Firefox bugs, and a Critical cyber rating, available only to Daybreak Red customers with hardware keys and IP allowlisting. It is not in Microsoft Foundry, so here is what an Azure team in Sweden can run instead: MDASH in the EU region, Mythos via Glasswing, and the patch math the exploit curve now demands.
SSHhole: the OpenSSH bypass and your Azure Alpine audit
OpenSSH 10.6p2, released 5 November 2026, closes CVE-2026-123158: on musl-based Linux without PAM, any password of 257 to 1,024 bytes logs in to any account whose shadow field is "*", which is the Alpine default for root and for the user in Alpine's own cloud images. Azure Linux and every endorsed distro are unaffected, so here is the audit for where it actually hides: Alpine VMs from VHDs, App Service custom containers built from Microsoft's SSH template, and sshd pods.
Brotli2Shell hits Node.js and .NET: your Azure patch plan
CVE-2026-111497 is an out-of-bounds write in Google's Brotli decoder that Node.js says permits remote code execution, and it reaches Node.js 22, 24 and 26, .NET 8, 9 and 10, and the PyPI Brotli package. Fixes landed on 21, 22 and 25 October 2026, Imperva has a working exploit on Node 24.22.0, Node 20 will never be patched, and this guide covers who patches what in Azure, which code paths decode untrusted Brotli, and a seven-step audit.
PHP2Shell on Azure: your App Service and WAF audit
PHP2Shell, CVE-2026-109618 and CVE-2026-109619, gives an unauthenticated attacker code execution on any PHP web server below 8.2.35, 8.3.36, 8.4.27 or 8.5.12, and was exploited before the 21 October 2026 patches shipped. Azure App Service on Linux patched its built-in runtimes the same day, Windows App Service PHP will never be patched, and the Front Door and Application Gateway mitigations are off by default. An audit for Azure teams.
LiteLLM 1.107.0 was malware: your Azure AI gateway audit
LiteLLM 1.107.0 shipped on 21 October 2026 with a credential stealer and remote-code loader that ran on every Python start, harvesting Azure, Kubernetes, SSH and AI-tool credentials plus the proxy database of provider keys. A two-hour audit for Azure teams running LiteLLM in front of Azure OpenAI: find the digests, hunt the indicators, rotate in the right order, and decide whether the gateway belongs in API Management.
The same MCP flaw at Google, JPMorgan and DINUM: your audit
Google, JPMorgan Chase, Weaviate, France's DINUM and an Indonesian city government have each fixed the same server-side request forgery flaw in their MCP servers, with Google's case scored CVSS 8.0 as CVE-2026-14540. Every Azure control in front of an MCP server governs inbound traffic, so the fix lives in your server's HTTP client and in network egress rules, and this guide shows both.