为通用智能体构建的治理机制
Governance by Construction for Generalist Agents
AI总结：
本文提出了一种模块化的政策-as-code层，用于在不微调模型的情况下，通过与通用大语言模型智能体结合，实现可预测、可审计且符合合规要求的行为，在复合工作流中无需为每个领域重新构建智能体。
摘要：
企业智能体日益被期望在多个工具和界面中自主运行，但生产部署需要通过构建来实施治理。系统必须指定哪些操作被允许、何时需要人类监督以及哪些信息可以暴露，而无需为每个领域重新构建智能体。本演示展示了CUGA的策略系统，这是一种模块化的策略-as-code层，能够与通用大语言模型智能体结合，以在复合工作流中实现可预测、可审计且符合合规要求的行为。我们提出了一种运行时治理架构，在执行的每一个关键阶段都强制执行策略干预。而不是被动地限制行为，策略在五个结构性检查点拦截智能体：规划上游（意图守卫）、在系统提示内引导推理（手册）、在工具调用边界处强制正确使用（工具指南）、在推理循环外作为人类在环的闸门用于高风险操作（工具批准）、以及在输出阶段过滤和结构化最终响应（输出格式器）。这些阶段将治理连续嵌入智能体的执行流程中，而不是将其视为事后考虑。通过一个医疗场景和多层次的执行干预，演示展示了动态手册注入用于结构化工具序列执行，意图守卫阻止恶意或意外有害请求，以及人类在环的工具批准检查点用于可能破坏性操作。该成果展示了类型化的治理原语如何加快、安全地部署企业智能体系统，同时提高政策遵守和执行一致性。
英文摘要：
Enterprise agents are increasingly expected to operate autonomously across tools and interfaces, yet production deployments require governance by construction. Systems must specify which actions are allowed, when human oversight is required, and what information may be exposed, without rebuilding the agent for each domain. This demo presents CUGA's policy system, a modular policy-as-code layer that composes with a generalist LLM agent to deliver predictable, auditable, and compliance-aware behavior in compound workflows without model fine-tuning. We present a runtime governance architecture that enforces policy interventions at every critical stage of execution. Rather than passively constraining behavior, policies intercept the agent at five structural checkpoints: upstream of planning (Intent Guard), within the system prompt to steer reasoning (Playbook), at the tool-call boundary to enforce proper usage (Tool Guide), outside the reasoning loop as a Human-in-the-Loop gate for high-risk actions (Tool Approvals), and at the output stage to filter and structure the final response (Output Formatter). Together, these stages embed governance continuously across the agent's execution pipeline rather than treating it as an afterthought. Using a healthcare scenario and a multi-layered enforcement intervention, the demo shows dynamic playbook injection for structured tool-sequence enforcement, intent guards that block malicious or accidental harmful requests, and human-in-the-loop tool approval checkpoints for potentially destructive actions. The artifact illustrates how typed governance primitives enable faster, safer deployment of enterprise agentic systems while improving policy adherence and execution consistency.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
