Vibe CodingCursor Long-Running Agents — 52-Hour Autonomous CodingAdwaitX·high signalXBlueskyLinkedInCopy linkCursor launches long-running agents for Ultra/Teams/Enterprise. 25-52+ hours autonomous operation, recursive planners spawning sub-planners and workers. 1000+ commits/hour in demos.SourceSource pageAdwaitX↳ Follow the threadPolicy dependency / Stack layerOrdewell uses a read-only coding agent as planner, then spawns per-task agent sessions gated by a marker-based verdict engineGitHubPolicy dependency / Stack layerCROSS-CATEGORY: Salesforce, Zendesk and Workable All Shipped Named Agent Portfolios on Sept 14, Each Metered in a Different UnitZendesk newsroom, Salesforce press release and Workable via GlobeNewswire (three independent Sept 14 announcements)Policy dependency / Stack layer'Do this as quickly as possible' repeatedly got a Claude session flagged by a corporate security directorr/ClaudeAIStack layer / ContrastEmergence World ran 10 agents per world for 16 days and found no frontier model contained an injected attack — one acted on poisoned memory 46 hours laterarXivStack layer / Update threadCROSS-CATEGORY: Three Vendors in 48 Hours Shipped Ways to Stop Paying Frontier-Model Prices — Salesforce's Own Model, TypeSafe's Non-LLM, Weave's RouterWeave OS blog, Salesforce press release and TypeSafe AI blog (three independent primary sources, Sept 15-16)Stack layer / Update threadn8n 2.40.0 preserves empty-text Anthropic thinking blocks across tool calls and enforces execution timeouts on stuck queue jobsGitHubStack layer / ContrastDifficulty-aware topology selection beats always-hierarchical multi-agent coding by 4.1 points at 40% of the costarXivPolicy dependency / Stack layerA Bun packaging quirk left a vulnerable undici in Cline after the CVE was supposedly remediatedGitHub