Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{191718,
        author = {Urooz Naqvi},
        title = {Bridging LLMs and Financial APIs: An MCP-Based Architecture for Automated Stock Trading},
        journal = {International Journal of Innovative Research in Technology},
        year = {},
        volume = {12},
        number = {no},
        pages = {50-60},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=191718},
        abstract = {This research addresses the challenges of connecting Large Language Models (LLMs) with financial APIs for automated data-driven trading. The study proposes a middleware design that improves security, reliability, and scalability for AI-driven trading systems. This study introduces a middleware framework based on the Model Context Protocol (MCP). This framework serves as a secure and expandable link between large language models and financial application programming interfaces, focusing on Zerodha. The implementation uses TypeScript to allow immediate trade processing based on instructions from language models.
Key components of the framework include fault management, user verification, and contextual oversight. The study looks at efficiency metrics such as response time, processing capacity, and successful trade completion rates in different operating scenarios to evaluate the overall performance of the proposed system.
The results indicate that the MCP-based middleware greatly enhances the safety, scalability, and efficiency of AI-driven trading systems. The structure reduces security risks and increases trade execution reliability, offering a practical way to integrate AI models into financial applications. This work provides useful insights for developers, researchers, and financial institutions by presenting a practical and secure method to implement AI-driven trading solutions. It links theoretical AI capabilities with real-world automated trading and highlights the potential for better trading strategies through the effective use of AI technologies.},
        keywords = {Large Language Models (LLMs), Algorithmic Trading, AI-Driven Trading Systems, Financial APIs, Middleware Architecture, Model Context Protocol (MCP)},
        month = {},
        }