Model Context Protocol (MCP) for AI Systems: A Survey of Architectures and Applications
Downloads
As Large Language Models (LLMs) continue to advance, the demand for standardized mechanisms that enable seamless, scalable, and context-aware integration of these models in a secure manner is increasing. To meet these requirements, the Model Context Protocol (MCP) has come into existence as an open interoperability standard that offers a common language for linking AI models with tools, resources, and external systems. The survey provides a complete overview of the architectural foundations, communication paradigm, client–server model, transport mechanisms, design principles, and primary application areas of MCP, such as enterprise systems, software engineering, healthcare, manufacturing, education, finance, and multi-agent systems. It also reviews the state of the art and research advances, outlining security issues and their mitigation, and evaluates the status of MCP-based implementations. MCP is shown to be a significant factor in improving interoperability, context-aware reasoning, workflow orchestration, and modular integration of AI, while also minimizing the need for proprietary interfaces. Moreover, it pinpoints key areas of research that are needed in the future, such as scalability, security, standardized benchmarking, interoperability, and long-term context management, offering researchers and practitioners working on the creation of next-generation AI systems significant directions for future research.
G. Qu, Q. Chen, W. Wei, Z. Lin, X. Chen, and K. Huang, “Mobile Edge Intelligence for Large Language Models: A Contemporary Survey,” IEEE Commun. Surv. Tutorials, vol. 27, no. 6, pp. 3820–3860, Dec. 2025, doi: 10.1109/COMST.2025.3527641.
A. Singh, A. Shetty, A. Ehtesham, S. Kumar, and T. T. Khoei, “A Survey of Large Language Model-Based Generative AI for Text-to-SQL: Benchmarks, Applications, Use Cases, and Challenges,” in 2025 IEEE 15th Annual Computing and Communication Workshop and Conference (CCWC), IEEE, Jan. 2025, pp. 00015–00021.
doi: 10.1109/CCWC62904.2025.10903689.
K. R. Kiran, “API Integration Barriers in Case-Based Process Management: A Review of Interoperability and Real-Time Response Constraints Outline,” Zenodo, vol. 25, no. 03, 2025,
doi: 10.5281/zenodo.16782616.
P. P. Ray, “A Survey on Model Context Protocol: Architecture, State-of-the-art, Challenges and Future Directions,” Apr. 18, 2025.
doi: 10.36227/techrxiv.174495492.22752319/v1.
H. Errico, J. Ngiam, and S. Sojan, “Securing the Model Context Protocol (MCP): Risks, Controls, and Governance,” 2025, doi: 10.48550/arXiv.2511.20920.
P. Venkiteela, “The New Interoperability Paradigm: Model Context Protocol (MCP), APIs, and the Future of Agentic AI,” Comput. Fraud Secur., pp. 1259–1271, May 2025, doi: 10.52710/cfs.817.
V. S. Narajala and I. Habler, “Enterprise-Grade Security for the Model Context Protocol (MCP): Frameworks and Mitigation Strategies,” in 2026 IEEE 5th International Conference on AI in Cybersecurity (ICAIC), IEEE, Feb. 2026, pp. 1–8.
doi: 10.1109/ICAIC67076.2026.11395723.
S. Singh, S. A. Pahune, P. Chatterjee, and R. Sura, “Advanced Machine Learning Techniques for Prediction of Customer Churn in Telecommunication Sector,” in 2025 IEEE 6th India Council International Subsections Conference (INDISCON), Rourkela, India: IEEE, 2025, pp. 1–6, August. doi: 10.1109/INDISCON66021.2025.11252233.
P. Chauhan, U. P. Rao, and M. M. Kailashchandra, “MCP-PUF: A Context-Hardened, ML-Assisted Authentication Protocol for Secure Healthcare IoT,” in 2025 IEEE 4th International Conference on Technology, Engineering, Management for Societal Impact using Marketing, Entrepreneurship and Talent (TEMSMET), IEEE, Oct. 2025, pp. 1–6.
doi: 10.1109/TEMSMET65536.2025.11467437.
A. Ehtesham, A. Singh, G. K. Gupta, and S. Kumar, “A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP),” pp. 1–21, 2025, doi: 10.48550/arXiv.2505.02279 Focus to learn more.
K. Jangiti, “Design and Validation of a Machine Identity Governance Framework for AI Agents in Multi-Cloud Environments,” in SoutheastCon 2026, Huntsville, AL, USA: IEEE, 2026, pp. 1–6, February- March.
doi: 10.1109/SoutheastCon63549.2026.11476363.
G. Tod-Raileanu, S.-D. Axinte, A.-M. Dincă, and I. C. Bacivarov, “Security Posture Evaluation of Model Context Protocol Servers Against Prompt Injection,” in 2025 IEEE 31st International Symposium for Design and Technology in Electronic Packaging (SIITME), IEEE, Oct. 2025, pp. 249–252. doi: 10.1109/SIITME67657.2025.11293682.
V. Ayyagari, “Model Context Protocol for Agentic AI: Enabling Contextual Interoperability Across Systems,” Int. J. Comput. Exp. Sci. Eng., vol. 11, no. 3, pp. 6072–6082, Aug. 2025,
doi: 10.22399/ijcesen.3678.
A. Shah, A. Chernova, E. Tomson, L. Porter, W. G. Griswold, and A. G. Soosai Raj, “Students’ Use of GitHub Copilot for Working with Large Code Bases,” in Proceedings of the 56th ACM Technical Symposium on Computer Science Education V. 1, New York, NY, USA: ACM, Feb. 2025, pp. 1050–1056. d
oi: 10.1145/3641554.3701800.
K. Chen, J. Keung, W. Li, X. Shao, Y. Li, and X. Ma, “Teaching Software Engineering with LLM and MCP Integration: From Classroom to Industry Practice,” no. 3, 2026, doi: 10.48550/arXiv.2606.19167.
M. Mittal, “AI in Healthcare: The Next Frontier In Medical Diagnosis,” Int. J. Inf. Technol. Manag. Inf. Syst., vol. 16, no. 2, pp. 1448–1455, Mar. 2025,
doi: 10.34218/IJITMIS_16_02_091.
M. R. C. Mukkolakkal, “Deploy, Calibrate, Monitor, Heal -- No Human Required: An Autonomous AI SRE Agent for Elasticsearch,” 2026.
doi: 10.48550/arXiv.2604.03933.
N. Krishnan, “Advancing Multi-Agent Systems Through Model Context Protocol: Architecture, Implementation, and Applications,” arXiv Prepr. arXiv2504.21030, Apr. 2025,
doi: 10.48550/arXiv.2504.21030.
Z. Zhou et al., “MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents,” 2026,
doi: 10.48550/arXiv.2602.14281.
S. P.Sivashanmugam, S. Kumara, A. Mohile, and D. R. Suram, “Improving Detection Accuracy of Phishing Attacks with Feature Selection and Machine Learning Techniques in Cybersecurity,” in 2026 1st International Conference on Emerging Trends in Advancements and Applications of Computational Intelligence Techniques (ETAACT), Bhubaneswar, India: IEEE, Apr. 2026, pp. 1–6.
doi: 10.1109/ETAACT69135.2026.11542138.
A. Fernandes et al., “Evaluating Model Context Protocol in the context of a Conversational Agent supporting an Agro-Food Blockchain Logistics Platform,” Procedia Comput. Sci., vol. 278, pp. 742–750, 2026, doi: 10.1016/j.procs.2026.03.046.
M. S. Farooq, M. Sarwar, N. A. Nawaz, U. Ahmed, M. H. Jamil, and J. Ahmad, “Advanced LLM-based Model-Context-Protocol and Agent-to-Agent Protocol for Child Medical Emergency,” in 2025 6th International Conference on Innovative Computing (ICIC), IEEE, Dec. 2025, pp. 1–6.
doi: 10.1109/ICIC68258.2025.11413209.
M. Gao, L. Li, X. Ming, and G. Shen, “Research on Standardized Encapsulation of Model Context Protocol Tools and AI Agent Integration,” in 2025 International Conference on Aerospace Information Perception and Intelligent Processing (AIPIP), IEEE, Oct. 2025, pp. 328–331.
doi: 10.1109/AIPIP66876.2025.11299220.
L. M. V. da Silva, A. Köcher, and F. Gehlhoff, “Beyond Formal Semantics for Capabilities and Skills: Model Context Protocol in Manufacturing,” in 2025 IEEE 30th International Conference on Emerging Technologies and Factory Automation (ETFA), IEEE, Sep. 2025, pp. 1–4.
doi: 10.1109/ETFA65518.2025.11205601.
V. Jadav, “Multi-Context Protocol (MCP) Hosts provide secure and real-time integration of AI tools,” in 2025 International Conference on Computer and Applications (ICCA), IEEE, Dec. 2025, pp. 1–6. doi: 10.1109/ICCA66035.2025.11430968.
A. Altınörs and B. T. Akgün, “A Model Context Protocol Application for Long-Term Memory Management,” in 2025 3rd International Conference on Foundation and Large Language Models (FLLM), IEEE, Nov. 2025, pp. 790–797.
doi: 10.1109/FLLM67465.2025.11391050.
