Chaotic Eclipse released YellowKey and GreenPlasma PoCs one day after May 2026 Patch Tuesday. A USB-borne FsTx folder plus a Ctrl-key reboot drops cmd.exe inside WinRE on a BitLocker-protected machine. Covers WinRE-only behavior, the CTFMON SYSTEM elevation path, the RedSun silent-patch dispute, and what defenders can actually do while unpatched.
May 12: TeamPCP open-sourced the Shai-Hulud worm on GitHub. Datadog mapped its module pipeline (Loader/Provider/Collector/Dispatcher/Sender/Mutator) and Claude Code SessionStart hook for persistence. May 15: BreachForums opened a paid attack challenge. Detection notes for the copycat wave.
What to patch, rotate, and grep after OpenClaw 2026.4.22. Walks CVE-2026-44112/44113/44115/44118 as one chain on agent runtime, with detection log fields and 24h/1w response steps.
Malicious node-ipc 9.1.6/9.2.3/12.0.1 fire on require(), not postinstall. 12.0.1 is SHA-256 gated (targeted), so a working app isn't safe. Exfils dev/CI secrets via DNS TXT.
Out-of-bounds read in Ollama's GGUF loader before 0.17.1. If your Ollama API is network-accessible, a crafted model file can exfiltrate env vars, API keys, system prompts, and conversation fragments from process memory.
Google extended Binary Transparency to its Android apps and Mainline modules starting May 2026. How the public log and verification tools differ from code signing, what's actually covered, and what the ADB-based verification workflow looks like for researchers.
Android's May 2026 bulletin patches CVE-2026-0073, a Wireless ADB auth bypass from mishandled EVP_PKEY_cmp return values. Adjacent network attackers bypass mutual TLS and get shell-level RCE on Android 14 through 16-qpr2. AOSP diff and impact breakdown included.
Next.js 16.2.6 / 15.5.18 dropped 13 security advisories at once. The impact depends on whether you use App Router, Middleware, RSC, or self-hosted Node.js server — here's where to look before upgrading.
CVE-2026-26268, fixed in Cursor 2.5, allowed AI agents to rewrite insufficiently protected .git config and Git hooks, leading to out-of-sandbox RCE on the next Git operation.
ZDI-26-305 discloses a sandbox bypass in OpenAI Codex. Processing a repository containing malicious JavaScript can lead to code execution under the user's privileges outside the sandbox.
How much does periodic password rotation or character-class enforcement actually help? A look at the numbers: leak probability, entropy, and user behavior.
