cs.CRNov 13, 2026

Securing Multi-Agent Systems by Design

Authors: Xilong Wang, Xiaoxing Chen, Neil Gong

Organizations: Duke University

Abstract

Multi-agent systems solve complex tasks by decomposing them into subtasks, delegating the subtasks to agents, and aggregating the agents' responses using an aggregation rule. Malicious agents can manipulate the final solution by providing malicious responses to their assigned subtasks. In this paper, we first propose response manipulation attacks (ReMA), which iteratively optimize malicious responses to maximize attack effectiveness. Specifically, we formulate untargeted and targeted attacks as optimization problems and develop an approach to approximately solve these problems by iteratively refining malicious responses based on feedback from a surrogate multi-agent system. To defend against such attacks, we further propose SecMAS, a secure aggregation rule that leverages the semantic similarities among agent responses to select the response that is most consistent with a sufficiently large subset of the other responses. We theoretically establish security guarantees for SecMAS against both untargeted and targeted attacks. Extensive evaluations demonstrate that existing aggregation rules are vulnerable to ReMA, whereas SecMAS effectively defends against all evaluated attacks, including adaptive variants of ReMA, while maintaining utility comparable to that of existing aggregation rules.

Figures & tables

Explore similar work

CardsList
  1. Architecture Matters for Multi-Agent Security

    Apr 25, 2026Ben Hagag, William L. Anderson, Christian Schroeder de Witt +1Multi-Agent SystemsMulti-Agent System Security

  2. GroupGuard: A Framework for Modeling and Defending Collusive Attacks in Multi-Agent Systems

    Mar 14, 2026Yiling Tao, Xinran Zheng, Shuo Yang +2Graph-Based Anomaly DetectionLLM Agent Security

  3. When Agents Go Rogue: Activation-Based Detection of Malicious Behaviors in Multi-Agent Systems

    Jul 7, 2026Haowen Xu, Xue Tan, Lei Ma +6LLM Agent SecurityAI Agent Monitoring