arXiv cs.AI· Nils Palumbo, Sarthak Choudhary, Jihye Choi, Guy Amir, Prasad Chalasani, Isil Dillig, Somesh Jha·· 3 天前AI 评分32
面向智能体系统的形式化策略执行方法
Formal Policy Enforcement for Agentic Systems
AI 导读
研究者提出一种基于运行时监控的智能体安全方法，允许用户用 Datalog 指定策略，并通过构建事件图在运行时执行。当智能体即将执行工具调用等动作时，该方法结合当前系统状态与历史事件判断是否违反策略，进而阻止或放行该动作。该方法已实现并在多个 benchmark 上评估，研究团队将其视为迈向可证明安全智能体系统的一步。
正文 · 原文
Abstract:As agentic computing systems grow increasingly popular, they raise serious security concerns due to their susceptibility to various attacks. While prior research efforts aim to mitigate these concerns, they are often too coarse-grained and offer no formal correctness guarantees. In this work, we present a novel approach to agentic security based on runtime monitoring. Our method allows the user to specify their intended policy in Datalog and enforces the specified policy at runtime by constructing event graphs. When an agent is about to execute an action such as a tool call, our approach determines whether the action would violate the policy in light of the current system state and the history of past events. The action is then blocked or permitted accordingly. We implemented our approach and evaluated it across multiple benchmarks, demonstrating its merits in real-world settings. We view this work as a step towards agentic systems that are provably safe and secure.
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2602.16708 [cs.CR] |
| (or arXiv:2602.16708v4 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2602.16708 arXiv-issued DOI via DataCite |
Submission history
From: Nils Palumbo [view email]
[v1]
Wed, 18 Feb 2026 18:57:12 UTC (437 KB)
[v2]
Thu, 19 Feb 2026 03:40:05 UTC (437 KB)
[v3]
Fri, 8 May 2026 15:45:27 UTC (83 KB)
[v4]
Mon, 12 Oct 2026 04:22:13 UTC (94 KB)
来源：arXiv cs.AI · arxiv.org