Cursor open-sources its official plugin marketplace cursor/plugins, standardizing the ecosystem with plugin.json + skills + MCP definitions (+470 stars in one day); apache/maka enters the Apache incubator with an append-only event log recording every tool call and permission decision for auditable local-first agent workbenches; magnitudedev/magnitude auto-detects hardware, downloads, and runs models locally out of the box for offline agents; vercel/eve puts agent capabilities into convention directories like tools/, skills/, and schedules/ — the filesystem is the interface. On the framework side, pydantic-ai ships a v2.32.1 patch.
ToolLIFT lifts tool trajectories to function-level workflow graphs and consistently beats SOTA on three OOD benchmarks; SkillTV-Bench uses 681 cases to show skill-aware judge skills boost agent evaluation accuracy by 14.8pp; TRIO-20's prespecified equivalence study finds zero unauthorized calls from GPT-5.6 across 840 trajectories, but higher reasoning effort increases rule-probing rate by 14.3pp
Three papers today converge on one core question: **are AI Agents production-ready?** The answer is unanimously — far from it. HANDBOOK.md reveals that even the strongest frontier models achieve only **36.2%** SOP compliance when dropped into a simulated enterprise; a LangGraph paper delivers three actionable stateful workflow recipes plus a decision guide on when *not* to use LangGraph; and MM-ToolSandBox is the first benchmark to quantify how hard visually-grounded tool calling really is — the best of 12 models still falls below 50% success. Three dimensions — compliance evaluation, framework design, visual tool use — together map out exactly how far Agents are from real-world deployment.
Today's common theme: **the way we evaluate agents is itself broken**. The first paper audits major tool-calling benchmarks and finds nearly 20% of scores are wrong; the second uses replay analysis to show which benchmarks can be stopped early for reliable conclusions (SWE-bench is the exception); the third introduces the first multimodal web agent benchmark that jointly evaluates task completion and guide generation — screenshot input, dual-objective scoring, and even the strongest models complete less than 40%. Read all three for a complete picture of the crisis in agent evaluation and where to go from here.
Three papers tackle production-grade agent reliability from different angles: MemCon models memory operations as an RL problem so agents learn when to store, retrieve, and forget — up to +15.2 points on 6 benchmarks; AgentCheck turns MCP servers into a debugging surface for reproducing tool faults and verifying fixes, filling a long-standing gap in the MCP ecosystem; AgentAbstain uses 263 paired tasks to show that even the strongest frontier models score below 60% on 'should-not-act' scenarios, and abstention ability barely correlates with task-solving ability — swapping in a stronger model won't fix this.
Three papers today converge on one theme — moving agents from experiments to reliable production: a multi-agent troubleshooting architecture deployed at hyperscale cloud with 90%+ autonomous resolution; a memory mechanism that lets agents learn from past tool-call successes and failures without retraining; and the first systematic comparison of six AI-assisted development process frameworks across six dimensions.