AI 中文总结
本文提出AuthorityLens框架，从系统权威、权威分离和主体权威三个维度衡量基于LLM的智能体系统的实际权威结构，并通过Codex、OpenCode和Gemini CLI的13种配置实验，揭示名义配置与实现权威之间的差异。
AI 中文摘要
基于LLM的智能体越来越多地被部署在真实系统中，对重要资源和决策拥有权威。这些智能体通常与持有不同形式权威的人类和基于LLM的参与者一起运作。然而，工作流角色、权限设置和审查机制并不一定反映实践中实现的权威。我们引入了AuthorityLens，一个用于衡量系统权威结构的框架。从权威组合出发，我们沿着三个维度评估系统：系统被授权做什么（系统权威）、行使该权威所需的联合参与程度（权威分离）、以及每个参与者持有的权威程度（主体权威）。我们从足以在允许的运行时状态下实现每个结果的最小参与者组合中推导出这些度量。我们将AuthorityLens应用于Codex、OpenCode和Gemini CLI，在13种操作配置下，覆盖一个共同的智能体操作组合。我们发现名义配置并不能清晰地映射到实现的权威上。在Codex中，完全访问仅略微改变系统权威，但显著地将权威集中在执行的助手身上。OpenCode的构建和计划配置具有相同的系统权威和权威分离，尽管工作流和根级权限不同。在Gemini CLI中，基于模型的审查增加了权威分离，而不改变系统权威。主体权威进一步区分了权威复制和权威分离：生成的或委派的智能体可以成为相同权威的替代持有者，而不增加所需的联合参与。总之，这些结果表明AuthorityLens提供了一个统一的框架，用于衡量和比较不同智能体系统间实现的权威结构。
英文摘要
LLM-based agents are increasingly deployed with authority over consequential resources and decisions in real systems. These agents often operate alongside human and LLM-based participants who hold different forms of authority. Yet workflow roles, permission settings, and review mechanisms do not necessarily reflect the authority realized in practice. We introduce AuthorityLens, a framework for measuring a system's authority structure. Starting from an authority portfolio, we evaluate a system along three dimensions: what the system is authorized to do (System Authority), how much joint participation is required to exercise that authority (Authority Separation), and how much authority each participant holds (Principal Authority). We derive these measurements from the minimal combinations of participants sufficient to realize each outcome across admissible runtime states. We apply AuthorityLens to Codex, OpenCode, and Gemini CLI across 13 operating configurations over a common portfolio of agent operations. We find that nominal configurations do not map cleanly onto realized authority. In Codex, Full Access changes System Authority only marginally while substantially concentrating authority in the executing Assistant. OpenCode's Build and Plan configurations have the same System Authority and Authority Separation despite different workflows and root-level permissions. In Gemini CLI, model-based review increases Authority Separation without changing System Authority. Principal Authority further distinguishes authority replication from authority separation: spawned or delegated agents can become alternative holders of the same authority without increasing the required joint participation. Together, these results demonstrate that AuthorityLens provides a unified framework for measuring and comparing realized authority structures across agent systems.
Comments76 pages, 5 figures, including appendices
