超越智能体运行框架：软件工厂编排的边界标准——单运营者案例研究
原标题：Beyond the Agent Harness: Boundary Criteria for Software-Factory Orchestration - A Single-Operator Case Study
AI 摘要
该论文提出一套边界判定标准，用于区分编码智能体运行框架（agent harness）、工厂编排层（factory harness）与领域层，并给出将故障归因拆分为缺陷来源、控制缺口与修复位置三个字段的编码方法。作者基于单一运营者服务六个项目的流水线及三个已发表系统的26个组件开发标准，冻结后由两个模型编码器对46个留出单元进行验证，机制层分类完全一致，事件字段一致率为18-19/20（κ 0.86-0.93），但存在较多信息缺失导致的歧义。修订版v1.1将机制分类与事件归因分离，敏感性重编码一致性有所下降（κ 0.78-0.93），作者强调一致性不等于准确性或实用性，且尚未测试人类编码器。
正文节选
Beyond the Agent Harness: Boundary Criteria for Software-Factory Orchestration A Single-Operator Case Study Abstract Several recent systems add a layer above the coding-agent harness that organises many agent runs into continuous delivery. The existence of such a layer is not our contribution. What is missing is a rule for deciding which layer a mechanism or a failure belongs to. We propose boundary criteria based on an operational definition of a run. A mechanism confined to one run belongs to