Is all security broken?

Is all security broken? This week on Agents Hour, Shane and Abhi dig into all security incidents from the past seven days. PHP announced two critical vulnerabilities rated 9.2, Railway disclosed a security incident, and major U.S. banks are finding unauthorized third-party AI agents operating inside customer accounts. On the defensive side, 1Password launched a platform for agents and Anthropic's Frontier Red Team shared how Claude can formally verify software at large scale. Also on the show: Anthropic filed its S-1 for an IPO and committed 5% of its shares to a new Public Benefit Fund. DeepSeek released V4.1-Flash and open-sourced DeepInfer, the inference framework behind it, along with revenue figures from its production clusters. More open-weight models arrived with Kimi K3.1 Pro, MiniMax M3.1 Flash and Meta's Muse Spark Open, while Kimi K3.1 and Qwen 4 are reportedly being tested in the arena under code names. Meanwhile, Claude Duet opened as a research preview, a new app where Claude comes to you with what needs your attention. 📚 MASTRA RESOURCES https://mastra.ai https://x.com/mastra_ai https://mastra.ai/community/discord https://github.com/mastra-ai Learn Mastra: https://mastra.ai/learn https://mastra.ai/books/principles-of-building-ai-agents https://mastra.ai/books/patterns-of-building-ai-agents WHAT IS MASTRA Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development, from prototype to production.