AuthorityLens: Rethinking LLM-Based Agent Systems Through the Lens of Authority

2026-09-26

AI Summary

This study addresses the discrepancy between nominal permissions and actual authority structures in LLM agent systems, for which no unified measurement framework exists. We propose AuthorityLens, a novel framework that introduces an authority measurement method based on minimal participating coalitions. Through formal modeling and runtime state analysis, it quantifies actual authority structures across system, separation, and subject dimensions. Experiments on mainstream CLI tools such as Codex demonstrate that this framework effectively reveals discrepancies between nominal configurations and actual authority, as well as agent replication phenomena. Furthermore, our findings indicate that full-access modes exacerbate power concentration, whereas model-level review enhances separation without altering overall system privileges.

Abstract

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.

Problem

• LLM-based agents
• authority structure
• permission settings
• realized authority
• agent systems

Innovation

• AuthorityLens
• LLM-based agents
• authority structure
• authority separation
• principal authority