Vibe CodingCursor 2.2 Multi-Agent Judging Auto-Picks Best Parallel RunCursor·high signalCursor auto-evaluates parallel agent runs and recommends the best. New Debug Mode with runtime instrumentation. Plan Mode gets Mermaid diagrams.SourceSource pageCursor↳ Follow the threadStack layer / Threat patternVaronis discloses 'Rogue Agent' — a Dialogflow CX flaw letting one permission poison a shared agent runtimeNxCodeStack layer / Threat patternTip: Update Headless Claude Code to v2.1.207 — Managed-Settings Consent Was Silently Auto-RecordedClaude Code ChangelogStack layer / Threat patternFriendly Fire: auto-approve coding agents run attacker code during security scansSecurityWeekPolicy dependency / Stack layerFirst Recon AI ships AI Security Runtime GA — inline policy enforcement on every agent-to-agent and agent-to-tool callBusiness WirePolicy dependency / Stack layerCodenotary ships AgentMon 3 for agent runtime policy and compliance loggingAI Agent StoreStack layerCursor 3.11 Ships Side Chats and Cloud Agent Hooks (beforeSubmitPrompt, subagentStart, stop)Cursor ChangelogStack layerTip: Use Cursor's New Cloud Agent Hooks to Build Self-Correcting LoopsCursor ChangelogStack layer / Update threadChatGPT Work vs Claude Cowork: OpenAI Undercuts Anthropic With a Free-Tier Autonomous Work Agent and 1,400 ConnectorsAllwork.space / Bloomberg