MCP-Driven Pathways for Intelligent Reference Services in Libraries

Abstract:  [Purpose/Significance] To address the issues of insufficient intelligence, passive service models, and inefficient resource integration in library reference services, this study explores intelligent transformation pathways driven by the Model Context Protocol (MCP) service. The goal is to leverage standardized protocols to break down barriers between large language models (LLMs) and heterogeneous library resource systems, thereby constructing a new paradigm of proactive, precise, and personalized smart services, and providing both theoretical and practical solutions for the development of smart libraries. [Methods/Process] Based on an analysis of the technical features of MCP and its compatibility with library business logic, a four-layer interaction model of "Resource-MCP-LLM-User" is constructed. A layered service system architecture is designed, comprising the MCP service layer, MCP protocol layer, foundational LLM layer, and user layer. Through protocol parsing, tool encapsulation, and context management mechanisms, dynamic scheduling and knowledge integration of multi-source resources by LLMs are achieved. [Results/Conclusions] The study finds that MCP services can dynamically extend the boundaries of intelligent reference services according to library capabilities, providing a scalable theoretical framework and technical implementation guide for the intelligent transformation of libraries, and promoting the evolution of public cultural services towards proactivity and knowledge orientation.