Covers AI application security, model risk, and agentic system abuse.
vLLM 0.31.0 adds a preload daemon and CRIU snapshots, moves the default build to CUDA 13.0, and removes or renames flags. What to check before upgrading.
· 3 min
Ollama 0.40 runs supported models on MLX by default on Apple Silicon, and 0.40.2 keeps a backup of every upgraded model on disk. How to clean up.
· 4 min
FloorGen is a free floor plan maker with no sign-up: 2D, 3D, code checks and exports, plus an MCP server that lets Claude or ChatGPT edit plans safely.
· 7 min
x64dbg-MCP Server before 1.1 exposed every debugger tool over HTTP, unauthenticated, on all interfaces. CVE-2026-107824 scores 9.3; no exploitation.
· 6 min
CVE-2026-103435 (CVSS v4.0 7.7) is a symlink race in Claude Code before 2.1.129, fixed in May 2026, that can redirect a write outside the project. No exploitation reported.
· 4 min
CVE-2026-104850 lets a malicious MCP server make the TypeScript SDK's OAuth client send refresh tokens and client secrets to an attacker's server. Fixed in 1.31.0 / 2.2.0.
· 4 min
Google's GTIG logged 141 exploited flaws in Jan–Aug 2026, more than all of 2025, and lists LiteLLM CVE-2026-42271 and Langflow CVE-2026-5027 as exploited.
· 8 min
vLLM v0.30.0 adds DeepSeek-V4.1-Flash, GLM-5.3-Flash, Fast Start and HiSparse, and changes defaults that can break an upgrade. What to check.
· 4 min