LLM-based agents are increasingly deployed in workflows where generated outputs may directly trigger state-changing actions. This creates an execution-boundary problem: proposed actions must be governed before they are executed. We study this problem through economically consequential multi-agent interactions and argue that deployment-grade agent systems should separate proposal generation from environment-facing execution. To operationalize this principle, we introduce the Organizational Control Layer (OCL), a model-agnostic governance infrastructure that intercepts generated actions before execution through policy enforcement and escalation, without modifying the underlying LLM generator. We evaluate OCL on adversarial buyer--seller negotiation environments adapted from AgenticPay. Across multiple frontier LLM backends, OCL reduces unsafe executions from 88% to near-zero while increasing valid success from 12% to 96%. Results further reveal a safety--utility tradeoff: strict governance improves compliance and reliability against policy and constraint violations, but can reduce flexibility in tightly constrained markets. These findings suggest that deployment-grade LLM agent systems require explicit governance at the boundary between language generation and executable actions. The source code is available at: https://github.com/SHITIANYU-hue/amai_ocl

翻译：基于大语言模型的代理正越来越多地部署在工作流程中，其生成的输出可能直接触发改变状态的行动。这产生了一个执行边界问题：在执行所提议的行动之前必须对其进行治理。我们通过具有经济后果的多代理交互来研究这一挑战，并论证了可部署的代理系统应将提议生成与环境交互的执行相分离。为实现这一原则，我们提出了组织控制层（OCL）——一种与模型无关的治理基础设施，它通过策略执行和升级机制在生成行动被执行前进行拦截，无需修改底层的大语言模型生成器。我们在改编自AgenticPay的对抗性买方-卖方谈判环境中评估了OCL。在多个前沿大语言模型后端上，OCL将不安全执行率从88%降至接近零，同时将有效成功率从12%提升至96%。结果进一步揭示了安全性与有效性的权衡：严格的治理能提升对策略和约束违规的合规性与可靠性，但在严格受限的市场中可能降低灵活性。这些发现表明，可部署的大语言模型代理系统需要在语言生成与可执行行动之间的边界上进行显式治理。源代码获取地址：https://github.com/SHITIANYU-hue/amai_ocl