From Tool Calling to MCP: Building a Natural Language Search with Spring AI and MCP Server

2026-05-19 · Source: NLP on Medium · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning · Depth: Intermediate, medium

Summary

This article details refactoring a natural language order search feature from a tightly coupled Spring AI tool-calling implementation to one utilizing the Model Context Protocol (MCP). MCP, an open standard by Anthropic, enables AI models to communicate with external tools and data sources via a universal interface. The refactored architecture separates the AI application (MCP client) from the data access layer (MCP server), allowing the `search_customer_orders` tool to be exposed as a standalone, independently deployable service. The MCP server, built with Spring Boot and `spring-ai-starter-mcp-server-webmvc`, registers the `CustomerOrderTool` as a `ToolCallbackProvider`. The AI application, using `spring-ai-starter-mcp-client` and `spring-ai-starter-model-openai`, connects to the MCP server via SSE, discovers its capabilities, and invokes tools through a `ChatClient` configured with `ToolCallbackProvider`.

Key takeaway

For AI Architects designing scalable, multi-application AI systems, adopting the Model Context Protocol (MCP) is crucial. This approach allows you to decouple AI applications from specific tool implementations, fostering reusability and independent versioning of business logic. Consider refactoring single-application tool-calling patterns to MCP servers when your capabilities need to be shared or managed with distinct release cycles, enhancing maintainability and platform growth.

Key insights

MCP standardizes AI model communication with external tools, enabling reusability and independent deployment.

Principles

• Separate AI orchestration from business logic.
• Standardize tool interfaces for broad compatibility.

Method

Expose data access capabilities as an MCP server using `spring-ai-starter-mcp-server-webmvc`, then configure an AI application as an MCP client with `spring-ai-starter-mcp-client` to discover and invoke these tools.

In practice

• Use MCP for shared AI capabilities.
• Deploy tools independently from AI applications.
• Centralize observability for tool invocations.

Topics

• Model Context Protocol
• Spring AI
• Tool Calling
• Natural Language Search
• MCP Server

Code references

• Mohsin0568/spring-ai

Best for: AI Engineer, AI Architect, Software Engineer

Related on AIssential

• The Protocol That Cleaned Up Our Agent Architecture — The Model Context Protocol (MCP) centralizes AI agent tool management, decoupling tool definitions from orchestrators for improved interoperability and scalability.
• Model Context Protocol Emerges as a Common Framework for Enterprise AI Systems — Model Context Protocol (MCP) standardizes AI model interaction with external tools and data, reducing vendor lock-in and improving enterprise AI system reliability.
• Bridging Design Systems and Code with MCP and AI Agents — Design systems, context engineering, and agentic AI combine to enable AIs to build consistent software by following explicit rules via MCP.
• Building Local AI Systems: Qwen3.6 + MCPs — The Model Context Protocol (MCP) enables Qwen3.6-35B-A3B to perform complex agentic tasks locally through universal tool integration.
• Making AI Personal: Building Context-Rich AI Tools with MCP • Kris Jenkins • GOTO 2025 — Agentic AI combines LLMs with programmer-defined tools, enabling context-rich answers to ambiguous personal queries.
• From Silos to Symphony: The Model Context Protocol and the Dawn of the Unified Machine — The Model Context Protocol (MCP) unifies AI model interaction with external tools and data through an open, standardized interface.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by NLP on Medium.