arXiv cs.AI· Nicolas Gertler, Claudio Novelli, Luciano Floridi·· 1 天前AI 评分31
为 AI 智能体运行框架（Agent Harness）建立可移植的规范描述
Towards Portable Specification of the Agent Harness
AI 导读
研究者将 LLM 智能体的运行框架（harness）拆解为八个功能组件，并提出一套可移植的规范框架，用于记录"哪个模型在哪个 harness 中运行"。同一模型在不同 harness 配置下，报告分数可相差数倍，且智能体的攻击面很大程度存在于 harness 中。
正文 · 原文
Abstract:Large language models act only through the infrastructure that surrounds them. That infrastructure, the harness, determines what a model can accomplish and how it fails, yet it has no settled definition. To address this gap, we decompose the harness into eight functional components that can be specified and compared across systems, and we propose a portable specification framework that records which model ran in which harness. The record is needed because different harness configurations can yield several-fold differences in reported scores for the same model, and because much of an agent's attack surface lies in its harness. Existing model cards and agent cards provide release and operational information, but do not establish the sequence of effective configurations within a run. Harnesses can change between releases and between runs, and, where agents edit their own tools and configuration, within a run. Enacted AI laws contain documentation duties relevant to covered systems, subject to their respective scopes and application dates, but do not prescribe the common harness record proposed here. Therefore, the harness has to be explicitly described and specified as a versioned part of the system.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.40917 [cs.AI] |
| (or arXiv:2610.40917v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.40917 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Nicolas Gertler [view email]
[v1]
Thu, 29 Oct 2026 19:43:14 UTC (769 KB)
来源：arXiv cs.AI · arxiv.org