Model Context Protocol (MCP): A Standard Interface for Tool-Aware AI Systems

Abstract: The rapid evolution of Large Language Models (LLMs) has created a pressing need for standardised mechanisms through which AI agents can interact with external tools, data sources, and services. Current integration approaches rely on ad-hoc REST wrappers or vendor-specific plugin frameworks, producing fragile, non-interoperable systems. This paper presents a comprehensive study of the Model Context Protocol (MCP), an open standard introduced by Anthropic in November 2024 that defines a uniform client-server interface for tool-aware AI systems. We examine the MCP architecture and analyse how it enables dynamic tool discovery, structured resource access, and secure OAuth 2.1 authentication. Through a literature survey of six peer-reviewed and pre- print studies covering MCP security threats, real-world server deployments, adaptive transport applications, and protocol-agnostic integration gaps, we identify current limitations and open research challenges. We further propose a novel MCP Bridge architecture designed to address the multi-protocol adoption gap. Results of our analysis suggest that MCP represents a foundational shift toward composable, auditable, and vendor-neutral AI tool integration.

Keywords: Model Context Protocol, Large Language Models, Tool-Aware AI, JSON-RPC 2.0, Agentic Systems, OAuth 2.1, MCP Bridge, API Integration