Vulnerability benchmarks rot. Cases leak into training data, scores measure memorization. We built N-Day-Bench: tests LLMs on finding real vulnerabilities in real repos, refreshed monthly from live GitHub advisories. Blinded judging. All traces public. Very interestingly, the latest model from @Zai_org, GLM 5.1 performs really well! Link: ndaybench.winfunc.com

🚨 CVE-2026-21636 in Node.js (@nodejs) Node.js permission model bypass via unchecked Unix Domain Socket connections (UDS) This vulnerability was autonomously discovered by winfunc.com, an AI agent that can find, exploit, and patch security vulnerabilities in codebases. Thanks to @_rafaelgss for triaging and fixing the issue.

everyone wins @winfunction.

Announcing Opcode Everyone's favorite Claude Code GUI Claudia is now Opcode... with a complete revamp. And we just hit 15K+ stars on GitHub! The support from the community has been immense! ❤️ What's new: - Feature parity with all new Claude Code features in the CLI. - A complete UI revamp, faster and cleaner. What's coming: - Opcode's own coding agent that exclusively uses open models while achieving SOTA performance on coding benchmarks. We're currently piloting this at companies requiring fully local on-prem coding agents. If you're one of them, reach out to us and we'll set this up for you! Try Opcode at opcode [dot] sh Like/retweet and reply to this tweet with "join" to get an invite to the Discord server and beta test our coding agent for free.