Hybrid Analysis for Secure MCP Tool Use in LLM Agents

Abstract

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 defens

Categories

Framework mappings

Suggested from the entry's categories.

@misc{he2026hybrid,
  title = {{Hybrid Analysis for Secure MCP Tool Use in LLM Agents}},
  author = {Ping He and Yuexiang Xie and Yaliang Li and Shouling Ji},
  year = {2026},
  month = jul,
  eprint = {2607.25297},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2607.25297}
}