A Survey on Model Context Protocol (MCP) in System Calls

Abhijeet Cholke; Vinodpuri Gosavi; Rohini Patil; Mandar Mokashi; Arati Kale

doi:10.2991/978-94-6239-723-1_14

<Previous Article In Volume

Next Article In Volume>

A Survey on Model Context Protocol (MCP) in System Calls

Authors

Abhijeet Cholke¹^{, *}, Vinodpuri Gosavi¹, Rohini Patil², Mandar Mokashi¹, Arati Kale¹

¹School of Computing, MIT Art, Design and Technology University, Kadamwak Wasti, India

²Department of Computer Science & Engineering, Bharati Vidyapeeth (Deemed to be University) College of Engineering, Pune, India

^*Corresponding author. Email: abhijeet.cholke@gmail.com

Corresponding Author

Abhijeet Cholke

Available Online 14 July 2026.

DOI: 10.2991/978-94-6239-723-1_14 How to use a DOI?
Keywords: Model Context Protocol (MCP); LLM Agents; System Calls Analogy; AI Integration Protocols; Research Maturity Models
Abstract: Recently, a new standardized interface called the Model Context Protocol (MCP) has been established which allows reliable interfacing of large language models (LLMs) with the tools and external resources. Due to its structured architecture, it can be analogously termed to system calls in operating systems, since both work as controlled intermediatory between the low-level execution environments and user-level requests. This survey will provide a thorough overview of the literature on MCP with the focus on its system call analogy, security issues, benchmarking and its usage on AI based workflows. We categorize the existing work into four major classifications, namely the security/safety aspect, the MCP server and benchmarking aspect, the investigation on the tool/function calling aspect and the system level observability aspect. In addition to this, we also indicate the open research challenges like the scalability aspect, the interoperability aspect and the need for formal verification. Finally, we remark on the future prospects of the research where MCP should be regarded as an `“syscall layer” for artificial intelligence” serving as the basic building block of the AI-native operating systems and also the multi-agent platforms.
Copyright: © 2026 The Author(s)
Open Access: Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume

Next Article In Volume>

Volume Title: Proceedings of the International Conference on Responsible, Risk-aware, and Regulated AI (RRRAI 2026)
Series: Advances in Intelligent Systems Research
Publication Date: 14 July 2026
ISBN: 978-94-6239-723-1
ISSN: 1951-6851
DOI: 10.2991/978-94-6239-723-1_14 How to use a DOI?
Open Access: Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

ris enw bib

TY  - CONF
AU  - Abhijeet Cholke
AU  - Vinodpuri Gosavi
AU  - Rohini Patil
AU  - Mandar Mokashi
AU  - Arati Kale
PY  - 2026
DA  - 2026/07/14
TI  - A Survey on Model Context Protocol (MCP) in System Calls
BT  - Proceedings of the International Conference on Responsible, Risk-aware, and Regulated AI (RRRAI 2026)
PB  - Atlantis Press
SP  - 149
EP  - 160
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-723-1_14
DO  - 10.2991/978-94-6239-723-1_14
ID  - Cholke2026
ER  -

download .riscopy to clipboard