Search papers, labs, and topics across Lattice.
This paper introduces MTGuard, a hybrid analysis framework that enhances the security of Model Context Protocol (MCP) tool use in large language model (LLM) agents by combining static and dynamic analysis techniques. The motivation behind this work is to address the safety risks associated with LLM agents being manipulated into executing malicious actions via MCP tools, a concern that existing static-only defenses fail to adequately mitigate. Evaluation results show that MTGuard significantly reduces harmful tool use while preserving the performance of LLM agents on legitimate tasks.
MTGuard effectively reduces harmful tool use in LLM agents, combining static and dynamic analysis to enhance security without sacrificing performance.
The rapid development of large language model (LLM) agents has enabled their broad adoption across diverse real-world tasks. To standardize interactions between LLM agents and external environments, Model Context Protocol (MCP) tools have emerged as a de facto standard and have been widely integrated into these systems. However, the use of MCP tools also introduces new safety risks, as LLM agents can be induced to perform malicious or unauthorized actions. Although prior work has proposed defenses for securing tool use in LLM agents, most methods rely on static analysis, i.e., inspecting prompts and generated outputs, which limits the defense effectiveness and robustness. To address these limitations, we propose MTGuard, a hybrid analysis-based defense framework designed to safeguard the use of MCP tools in LLM agents by leveraging lifecycle-aware static-dynamic co-analysis. Extensive evaluation demonstrates that MTGuard effectively mitigates multiple categories of harmful tool use across different LLM agents while maintaining performance on benign user tasks.
DAMO