AI Automates a Spring Boot System, Leaving Colleagues Stunned

This demo shows how a large language model can be used as an autonomous AI Agent to operate a Spring Boot application.

Core Idea

By annotating regular Spring Boot methods with @Tool, the methods become AI‑accessible tools. When a user sends a single natural‑language command such as “Help me create a MacBook Pro order and notify the warehouse”, the AI parses the intent, decides which tools to call, and executes them in the correct order.

Step‑by‑Step Setup

Project creation : Use JDK 17 and Spring Boot 3.3.x. Add the following Maven dependencies:

<project xmlns="http://maven.apache.org/POM/4.0.0">
  <modelVersion>4.0.0</modelVersion>
  <groupId>com.demo</groupId>
  <artifactId>spring-ai-agent-demo</artifactId>
  <version>1.0</version>
  <parent>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-parent</artifactId>
    <version>3.3.0</version>
  </parent>
  <properties>
    <java.version>17</java.version>
    <spring-ai.version>1.0.0</spring-ai.version>
  </properties>
  <dependencies>
    <!-- Spring Boot -->
    <dependency>
      <groupId>org.springframework.boot</groupId>
      <artifactId>spring-boot-starter-web</artifactId>
    </dependency>
    <!-- Spring AI -->
    <dependency>
      <groupId>org.springframework.ai</groupId>
      <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
      <version>${spring-ai.version}</version>
    </dependency>
  </dependencies>
</project>

AI model configuration : Add the model settings to application.yml.

server:
  port: 8080

spring:
  ai:
    openai:
      api-key: YOUR_API_KEY
      base-url: https://api.openai.com
      chat:
        options:
          model: gpt-4o

Alternatively, use Claude via OpenRouter:

spring:
  ai:
    openai:
      api-key: sk-xxxx
      base-url: https://openrouter.ai/api/v1
      chat:
        options:
          model: anthropic/claude-3.7-sonnet

Define Tools : Three Java classes are annotated with @Tool to expose them to the AI.

package com.demo.agent.tool;

import org.springframework.ai.tool.annotation.Tool;
import org.springframework.stereotype.Component;

@Component
public class StockTools {
  @Tool(description = "查询商品库存")
  public Integer queryStock(String productName) {
    System.out.println("[Tool] 查询库存：" + productName);
    return 99;
  }
}

package com.demo.agent.tool;

import org.springframework.ai.tool.annotation.Tool;
import org.springframework.stereotype.Component;

@Component
public class OrderTools {
  @Tool(description = "创建订单")
  public String createOrder(String productName) {
    System.out.println("[Tool] 创建订单：" + productName);
    return "订单创建成功";
  }
}

package com.demo.agent.tool;

import org.springframework.ai.tool.annotation.Tool;
import org.springframework.stereotype.Component;

@Component
public class WarehouseTools {
  @Tool(description = "通知仓库发货")
  public String sendWarehouseNotice(String productName) {
    System.out.println("[Tool] 通知仓库发货：" + productName);
    return "仓库已收到发货通知";
  }
}

AI Agent Controller : Wire the tools into a ChatClient and expose a /chat endpoint.

package com.demo.agent.controller;

import com.demo.agent.tool.OrderTools;
import com.demo.agent.tool.StockTools;
import com.demo.agent.tool.WarehouseTools;
import org.springframework.ai.chat.client.ChatClient;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RestController;

@RestController
public class AgentController {
  private final ChatClient chatClient;

  public AgentController(ChatClient.Builder builder, OrderTools orderTools, StockTools stockTools, WarehouseTools warehouseTools) {
    this.chatClient = builder.defaultTools(orderTools, stockTools, warehouseTools).build();
  }

  @GetMapping("/chat")
  public String chat(String message) {
    return chatClient.prompt().user(message).call().content();
  }
}

Run the application with mvn spring-boot:run (or via an IDE). The console shows Tomcat started on port 8080.

Demo Execution

Send a request:

http://localhost:8080/chat?message=帮我购买一台MacBook%20Pro，并安排发货

The AI automatically performs the workflow:

[Tool] 查询库存：MacBook Pro
[Tool] 创建订单：MacBook Pro
[Tool] 通知仓库发货：MacBook Pro

And returns:

MacBook Pro 库存充足。
订单已经创建成功。
仓库已经收到发货通知。

Why Multiple Tools Are Invoked

The model first infers the user’s goal (purchase a product), then determines the necessary sub‑tasks: check stock, create an order, and notify the warehouse. It sequences the calls as queryStock() → createOrder() → sendWarehouseNotice(), which is a true AI‑driven workflow rather than a hard‑coded if‑else chain.

Future Extensions

While the demo uses three simple tools, the same pattern can be expanded to cover full enterprise micro‑service ecosystems (user service, payment service, ERP, CRM, etc.). The AI Agent could automatically orchestrate refund processing, approval flows, work‑order routing, and even generate SQL statements.

Takeaway

This example illustrates the shift from chat‑only AI to AI that can execute real business workflows, positioning Spring Boot as a foundational platform for the emerging AI Agent era.