MCP-Secure: A Runtime Access Control Layer for Privilege-Aware LLM Agent Tooling
Top Cited Papers
Open Access
- 12 February 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Computer Society
- Vol. 7, 504-513
Abstract
Tool-enabled language-model agents introduce new security risks because their behavior evolves over multi-step workflows, yet existing defenses primarily rely on static allowlists or infrastructure isolation. This paper presents MCP-Secure, a lightweight, host-side enforcement layer for the Model Context Protocol (MCP) that applies scoped access, read-only defaults, and approval-gated privilege elevation at runtime. MCP-Secure tracks permissions through a session-level state machine and mediates every tool invocation without modifying agents or MCP servers. We evaluate the framework across 1,080 executions spanning multiple models, tasks, and adversarial simulations. Results show that scoped access alone blocks most unsafe behaviors, read-only enforcement reliably neutralizes all mutating attack vectors, and approval-gated elevation maintains strong safety while enabling controlled write operations. Across configurations, the wrapper also shapes agent planning, reducing unsafe attempts as policies tighten. These findings demonstrate that MCP-Secure provides a practical, reproducible mechanism for enforcing least-privilege constraints in tool-enabled LLM systems, offering strong adversarial resistance with interpretable safety-utility trade-offs.Keywords
This publication has 6 references indexed in Scilit:
- Enterprise-Grade Security for the Model Context Protocol (MCP): Frameworks and Mitigation Strategies2026
- Model Context Protocol (MCP): Landscape, Security Threats, and Future Research DirectionsACM Transactions on Software Engineering and Methodology, 2026
- Large Language Models in Cybersecurity: A Survey of Applications, Vulnerabilities, and Defense TechniquesAI, 2025
- LLM-Based Agents for Tool Learning: A SurveyData Science and Engineering, 2025
- MCP Guardian: A Security-First Layer for Safeguarding MCP-Based AI System2025
- A survey on large language model (LLM) security and privacy: The Good, The Bad, and The UglyHigh-Confidence Computing, 2024