AI-Powered Travel Planning with Model Context Protocol (MCP): A Survey

Migrating Large Language Models to Deal with Context Using a standardised protocol-oriented architecture, rather than traditional software methods, to build and combine AI appli- cations is an important advance. “TripSync”, a trip-planning web application, is used as an example of the use of the Model Context Protocol (MCP) to build modular, flexible, AI-based applications with good context handling. MCP is a client–server framework that provides a standard interface for connecting an LLM to other tools. The multiplexing of contexts onto inputs to the LLM enables multiple applications to use the same LLM with multiple data sources. At the same time, the introduction of this protocol must be accompanied by a concern about security and ecosystem problems. Although MCP solves many of the interoperability problems, it also creates new security, governance, and long- term maintenance challenges. Research has revealed quality- of-code challenges and vulnerabilities, such as tool poisoning across multiple MCP servers. By responding to these challenges with enhanced secu- rity measures, centralised registries, and community-dictated norms, developers can shape the next wave of AI systems that are more than just powerful; they’re also safe and dependable.

[1]     X. Hou, Y. Zhao, S. Wang, and H. Wang, ”Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions,” arXiv preprint arXiv:2503.23278v2 [cs.CR], 2025. [2]     A. Ehtesham, A. Singh, S. Kumar, and G. K. Gupta, ”A Survey of Agent Interoperability Protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP),” arXiv preprint arXiv:2505.02279v1, 2025. [3]     N. Krishnan, ”Advancing Multi-Agent Systems Through Model Con- text Protocol: Architecture, Implementation, and Applications,” arXiv preprint arXiv:2504.21030v1, 2025. [4]     M. M. Hasan et al., ”Model Context Protocol (MCP) at First Glance: Studying the Security and Maintainability of MCP Servers,” arXiv preprint arXiv:2506.13538v4, 2025. [5]     H. Guo et al., ”A Measurement Study of Model Context Protocol Ecosystem,” arXiv preprint arXiv:2509.25292v2, 2025. [6]     P. P. Ray, ”A Survey on Model Context Protocol: Architecture, State- of-the-art, Challenges and Future Directions,” TechRxiv, Apr. 2025. [7]     P. Pajo, ”Smithery.ai: A Model Context Protocol for Enhanced Large Language Model Integration and Cross-Industry Applications,” Re- searchGate, Mar. 2025. [8]     V. S. Narajala and I. Habler, ”Enterprise-Grade Security for the Model Context Protocol (MCP): Frameworks and Mitigation Strategies,” arXiv preprint arXiv:2504.08623v2, 2025. [9]     Q. Li and Y. Xie, ”From Glue-Code to Protocols: A Critical Analysis of A2A and MCP Integration for Scalable Agent Systems,” arXiv preprint arXiv:2505.03864v1, 2025. [10]    H. Li, Y. Xu, and T. Hong, ”EnergyPlus-MCP: A model-context-protocol server for ai-driven building energy modeling,” SoftwareX, vol. 32, p. 102367, 2025.