# Expose your Spring Boot API endpoints via MCP and Copilot **URL:** \ **Category:** Content Creators\ **Tags:** mcp, copilot, vscode, visual-studio-code\ **Created:** [14 August 2026 14:15 UTC](https://www.artofsm.art/t/expose-your-spring-boot-api-endpoints-via-mcp-and-copilot/22499 "2026-08-14T14:15:17Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![merefield](https://www.artofsm.art/user_avatar/www.artofsm.art/merefield/32/74_2.png) [@merefield](https://www.artofsm.art/u/merefield)\ **Post date:** [14 August 2026 14:15 UTC](https://www.artofsm.art/t/expose-your-spring-boot-api-endpoints-via-mcp-and-copilot/22499/1 "2026-08-14T14:15:17Z")
[![](https://www.artofsm.art/uploads/default/original/3X/b/a/ba2b641a5e4e9d454d060b6893d0db242c2f288a.jpeg "Expose your Spring Boot API endpoints via MCP and Copilot") ](https://www.youtube.com/watch?v=gzyU4yn0qGw) The video demonstrates how to expose Spring Boot API endpoints as callable tools for GitHub Copilot using the Model Context Protocol (MCP), enabling Copilot to invoke Java methods directly and enhance developer productivity within Visual Studio Code. It showcases a to-do list application where MCP annotations and configurations allow seamless integration, real-time synchronization, and practical interaction between Copilot and the Spring Boot service. ---
**Author:** ![merefield](https://www.artofsm.art/user_avatar/www.artofsm.art/merefield/32/74_2.png) [@merefield](https://www.artofsm.art/u/merefield)\ **Post date:** [14 August 2026 14:35 UTC](https://www.artofsm.art/t/expose-your-spring-boot-api-endpoints-via-mcp-and-copilot/22499/3 "2026-08-14T14:35:34Z")
The video demonstrates how to expose Spring Boot API endpoints as tools that GitHub Copilot can call directly, using the Model Context Protocol (MCP). This integration allows Copilot not only to write Java methods but also to invoke them, enhancing developer productivity. The setup is done within Visual Studio Code, requiring the Java Development Kit, Extension Pack for Java, and Spring Boot Extension Pack. The sample application used is a to-do list with existing operations, which are then exposed as MCP tools. In the Spring Boot application, the MCP server starter dependency is added via pom.xml, enabling the app to act as an MCP server. The to-do operations are annotated with MCP-specific annotations that define tool names, descriptions, and input parameters. These annotated methods delegate their logic to the existing to-do service, ensuring no duplication of business logic. The MCP server is configured to use a modern streamable HTTP transport in application.properties, which is necessary for Visual Studio Code to connect properly. Once the application is running, Visual Studio Code is configured to connect to the MCP endpoint through a JSON configuration file specifying the server URL and type. Starting the MCP client in VS Code discovers the five to-do tools exposed by the Spring Boot app. These tools are then made available to GitHub Copilot via the tools picker in Copilot chat, allowing Copilot to invoke the Java methods as needed. The video shows a practical example where Copilot is asked to add a new to-do item and then list all to-dos. Copilot calls the MCP tools in sequence, and the results are confirmed both in the chat response and on the web page, demonstrating that both the web interface and Copilot share the same underlying service and data repository. This confirms the seamless integration and real-time synchronization between the tools and the application. Finally, the demo concludes by shutting down both the MCP connection and the Spring Boot application, clearing the in-memory repository. The video highlights how Spring AI and MCP enable developers to turn their Java application methods into callable tools for Copilot, all managed within Visual Studio Code. Viewers are encouraged to consider which parts of their own Java applications could be exposed as Copilot tools, inviting engagement and further exploration.