Real Faults in Model Context Protocol (MCP) Software: a Comprehensive Taxonomy

TL;DR

This paper presents the first comprehensive taxonomy of real faults in Model Context Protocol (MCP) software, based on empirical data and practitioner surveys, to improve the robustness and security of MCP-based AI systems.

Contribution

It introduces a large-scale fault taxonomy for MCP servers and validates it through practitioner surveys, highlighting key fault categories and their practical relevance.

Findings

All identified fault categories occur in practice

Distinct characteristics differentiate MCP-specific faults from non-MCP faults

Insights inform development of more robust MCP-based systems

Abstract

The rapid adoption of foundation models has significantly expanded the capabilities of software systems, enabling them to perform complex language, reasoning, and interaction tasks that were previously difficult to automate. However, this progress has also introduced novel challenges that were largely absent in previous generations of software. In particular, the increasing integration of foundation models with external tools and resources raises new concerns regarding reliability, security, and robustness. The Model Context Protocol (MCP) has recently been proposed to standardize interactions between AI-based software systems, software tools, and external resources. Despite its growing adoption, there remains limited systematic understanding of real-world faults in MCP-based software systems. In this paper, we present the first large-scale taxonomy of faults in MCP servers, comprising five high-level fault categories derived from empirical evidence. To evaluate the completeness and generalizability of this taxonomy, we conduct a survey of MCP practitioners with diverse roles and experience levels. The results confirm that all identified fault categories occur in practice and reveal distinct characteristics that differentiate MCP-specific faults from non-MCP faults. Overall, this study provides actionable insights for researchers and practitioners by identifying the most error-prone and critical components of MCP-based systems. These insights can inform the development of more robust, reliable, and secure AI-enabled software systems that rely on MCP.