Bridging AI and Software Security: A Comparative Vulnerability Assessment of LLM Agent Deployment Paradigms

2025-07-08

AI Summary

Current LLM agent security research treats AI-specific vulnerabilities and traditional software flaws in isolation, lacking a unified cross-domain evaluation framework. Method: This work presents the first systematic comparison of two dominant deployment paradigms—Function Calling (FC) and Model Context Protocol (MCP)—and introduces a unified threat taxonomy integrating AI reasoning vulnerabilities with classical software security concepts. We empirically evaluate the robustness of seven LLMs against prompt injection, JSON injection, DoS, and multi-step chained attacks across 3,250 adversarial scenarios. Contribution/Results: FC exhibits higher overall attack success (73.5% vs. 62.59%), but risks concentrate at the system layer; MCP exacerbates LLM centralization and exposure. Chained attacks achieve 91–96% success rates, and advanced reasoning models show heightened exploitability. Our findings reveal how architectural choices fundamentally reshape the threat landscape, establishing novel conceptual insights and practical benchmarks for secure LLM agent design.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Safety and Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Large Language Model (LLM) agents face security vulnerabilities spanning AI-specific and traditional software domains, yet current research addresses these separately. This study bridges this gap through comparative evaluation of Function Calling architecture and Model Context Protocol (MCP) deployment paradigms using a unified threat classification framework. We tested 3,250 attack scenarios across seven language models, evaluating simple, composed, and chained attacks targeting both AI-specific threats (prompt injection) and software vulnerabilities (JSON injection, denial-of-service). Function Calling showed higher overall attack success rates (73.5% vs 62.59% for MCP), with greater system-centric vulnerability while MCP exhibited increased LLM-centric exposure. Attack complexity dramatically amplified effectiveness, with chained attacks achieving 91-96% success rates. Counterintuitively, advanced reasoning models demonstrated higher exploitability despite better threat detection. Results demonstrate that architectural choices fundamentally reshape threat landscapes. This work establishes methodological foundations for cross-domain LLM agent security assessment and provides evidence-based guidance for secure deployment. Code and experimental materials are available at https: // github. com/ theconsciouslab-ai/llm-agent-security.

Problem

• Compare security vulnerabilities in LLM agent deployment paradigms
• Assess AI-specific and software threats in 3250 attack scenarios
• Evaluate impact of architectural choices on LLM agent exploitability

Innovation

• Comparative evaluation of Function Calling and MCP paradigms
• Unified threat classification framework for AI and software vulnerabilities
• Tested 3,250 attack scenarios across seven language models