基于 Spring AI 构建多智能体协作系统（高级版）

SmartBrain2026-03-16 14:11

目录

一、项目背景

二、架构设计

[1. 整体架构（分层设计）](#1. 整体架构（分层设计）)

[2. 核心技术选型](#2. 核心技术选型)

三、工程结构

四、核心代码实现

[1. 依赖配置（pom.xml）](#1. 依赖配置（pom.xml）)

[2. 配置层](#2. 配置层)

[（1）OpenAI 配置（OpenAiConfig.java）](#（1）OpenAI 配置（OpenAiConfig.java）)

（2）支付宝配置（AlipayConfig.java）

[3. 数据模型（ShopAgentContext.java）](#3. 数据模型（ShopAgentContext.java）)

[4. 智能体服务层](#4. 智能体服务层)

（1）下单智能体（OrderAgentService.java）

（2）支付智能体（PaymentAgentService.java）

（3）物流智能体（LogisticsAgentService.java）

（4）交付智能体（DeliveryAgentService.java）

（5）客服智能体（CustomerServiceAgent.java）

[5. 流程编排（ShopWorkflowService.java）](#5. 流程编排（ShopWorkflowService.java）)

[6. 控制层（ShopAgentController.java）](#6. 控制层（ShopAgentController.java）)

[7. 启动类（ShopMultiAgentApplication.java）](#7. 启动类（ShopMultiAgentApplication.java）)

[8. 测试类（ShopAgentFlowTest.java）](#8. 测试类（ShopAgentFlowTest.java）)

五、配置文件（application.yml）

六、核心总结

七、运行说明

基于 Spring AI 框架构建购物场景多智能体协作系统，包括项目背景、架构设计、工程结构及核心代码实现，如依赖配置、配置层（OpenAI 配置和支付宝配置）、数据模型和智能体服务层等，旨在实现全链路智能化、低耦合、高扩展的购物流程管理，帮助数据服务提供商降本增效。

一、项目背景

在电商零售场景中，"下单 - 支付 - 物流 - 交付 - 客服" 是贯穿交易全生命周期的核心链路，传统单体系统存在三大痛点：

流程耦合度高：各环节强绑定，新增 / 修改业务规则（如物流渠道、支付方式）需重构整体逻辑；

智能化不足：固定规则难以适配个性化需求（如动态物流方案、智能客服应答）；

多工具集成复杂：OpenAI 大模型、支付宝 MCP、RAG 知识库等多源工具调用需大量胶水代码，维护成本高。

本项目基于 Spring AI 框架构建购物场景多智能体协作系统，将核心链路拆解为 5 个独立智能体，通过 StateGraph 实现流程编排，结合 OpenAI 原生接口、支付宝 MCP 工具、RAG 知识库，实现全链路智能化、低耦合、高扩展的购物流程管理，帮助数据服务提供商降本增效。

二、架构设计

1. 整体架构（分层设计）

|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| ┌─────────────────┐ │ Controller层 │ 对外暴露REST接口，接收请求并触发流程 ├─────────────────┤ │ Service层 │ 核心业务层：5个智能体+StateGraph流程编排 ├─────────────────┤ │ Config层 │ 配置OpenAI/支付宝客户端，依赖注入 ├─────────────────┤ │ DTO层 │ 定义上下文/请求模型，统一数据传递格式 └─────────────────┘ |

2. 核心技术选型

基础框架：Spring Boot 3.2 + Spring AI 1.0.0-M1

流程编排：Spring AI StateGraph

AI 能力：OpenAI 原生 API（Completions/Embeddings/Chat）

第三方工具：支付宝 MCP 工具

知识库：RAG 自定义 API + OpenAI Embeddings

三、工程结构

|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| src/main/java/com/shop/multiagent/ ├── ShopMultiAgentApplication.java // 启动类 ├── config/ // 配置层 │ ├── OpenAiConfig.java // OpenAI客户端配置 │ └── AlipayConfig.java // 支付宝MCP配置 ├── controller/ // 控制层 │ └── ShopAgentController.java // 购物流程入口接口 ├── service/ // 服务层 │ ├── agent/ // 5个智能体实现 │ │ ├── OrderAgentService.java // 下单智能体 │ │ ├── PaymentAgentService.java // 调支付智能体 │ │ ├── LogisticsAgentService.java// 查物流智能体 │ │ ├── DeliveryAgentService.java // 能交付智能体 │ │ └── CustomerServiceAgent.java // 做客服智能体 │ └── workflow/ // 流程编排 │ └── ShopWorkflowService.java // StateGraph流程组装 ├── dto/ // 数据模型 │ └── ShopAgentContext.java // 智能体上下文（传递数据） └── test/ // 测试类 └── ShopAgentFlowTest.java // 单节点+全流程测试 |

四、核心代码实现

1. 依赖配置（pom.xml）

|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <?xml version="1.0" encoding="UTF-8"?> <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 https://maven.apache.org/xsd/maven-4.0.0.xsd"> <modelVersion>4.0.0</modelVersion> <parent> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-parent</artifactId> <version>3.2.6</version> <relativePath/> </parent> <groupId>com.shop</groupId> <artifactId>multi-agent-demo</artifactId> <version>0.0.1-SNAPSHOT</version> <dependencies> <!-- Spring AI 核心（StateGraph+OpenAI） --> <dependency> <groupId>org.springframework.ai</groupId> <artifactId>spring-ai-openai-spring-boot-starter</artifactId> <version>1.0.0-M1</version> </dependency> <dependency> <groupId>org.springframework.ai</groupId> <artifactId>spring-ai-core</artifactId> <version>1.0.0-M1</version> </dependency> <!-- Spring Boot Web --> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-web</artifactId> </dependency> <!-- 支付宝MCP工具 --> <dependency> <groupId>com.alipay.sdk</groupId> <artifactId>alipay-sdk-java</artifactId> <version>4.38.0.ALL</version> </dependency> <!-- 测试依赖 --> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-test</artifactId> <scope>test</scope> </dependency> </dependencies> <repositories> <repository> <id>spring-milestones</id> <url>https://repo.spring.io/milestone\</url> </repository> </repositories> </project> |

2. 配置层
（1）OpenAI 配置（OpenAiConfig.java）

|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.config; import org.springframework.ai.openai.OpenAiApi; import org.springframework.ai.openai.OpenAiChatClient; import org.springframework.ai.openai.OpenAiCompletionsClient; import org.springframework.ai.openai.OpenAiEmbeddingClient; import org.springframework.beans.factory.annotation.Value; import org.springframework.context.annotation.Bean; import org.springframework.context.annotation.Configuration; @Configuration public class OpenAiConfig { @Value("${spring.ai.openai.api-key}") private String apiKey; @Bean public OpenAiApi openAiApi() { return new OpenAiApi("https://api.openai.com/v1", apiKey); } // 下单/交付智能体：流式Prompt输出 @Bean public OpenAiCompletionsClient completionsClient() { return new OpenAiCompletionsClient(openAiApi()); } // 客服智能体：RAG向量嵌入 @Bean public OpenAiEmbeddingClient embeddingClient() { return new OpenAiEmbeddingClient(openAiApi()); } // 通用对话（客服/物流） @Bean public OpenAiChatClient chatClient() { return new OpenAiChatClient(openAiApi()); } } |

（2）支付宝配置（AlipayConfig.java）

|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.config; import com.alipay.api.AlipayClient; import com.alipay.api.DefaultAlipayClient; import org.springframework.beans.factory.annotation.Value; import org.springframework.context.annotation.Bean; import org.springframework.context.annotation.Configuration; @Configuration public class AlipayConfig { @Value("{alipay.app-id}") private String appId; @Value("{alipay.private-key}") private String privateKey; @Value("{alipay.public-key}") private String alipayPublicKey; @Value("{alipay.gateway-url}") private String gatewayUrl; @Bean public AlipayClient alipayClient() { return new DefaultAlipayClient(gatewayUrl, appId, privateKey, "json", "UTF-8", alipayPublicKey, "RSA2"); } } |

3. 数据模型（ShopAgentContext.java）

|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.dto; import lombok.Data; // 智能体上下文：节点间数据传递载体 @Data public class ShopAgentContext { private String orderId; // 订单ID private String productInfo; // 商品信息 private String userId; // 用户ID private String tradeNo; // 支付宝交易号 // 各节点输出 private String orderInfo; // 下单智能体输出 private boolean paymentSuccess; // 支付智能体结果 private String logisticsInfo; // 物流智能体结果 private String deliveryInfo; // 交付智能体输出 private String customerServiceReply; // 客服智能体结果 } |

4. 智能体服务层
（1）下单智能体（OrderAgentService.java）

|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.service.agent; import com.shop.multiagent.dto.ShopAgentContext; import org.springframework.ai.openai.OpenAiCompletionsClient; import org.springframework.stereotype.Service; import reactor.core.publisher.Flux; @Service public class OrderAgentService { private final OpenAiCompletionsClient completionsClient; public OrderAgentService(OpenAiCompletionsClient completionsClient) { this.completionsClient = completionsClient; } // 下单节点：流式Prompt生成订单信息 public ShopAgentContext execute(ShopAgentContext context) { String prompt = String.format( "生成订单%s的JSON格式下单信息，用户%s，商品%s，包含：商品明细、金额、收货地址、下单时间", context.getOrderId(), context.getUserId(), context.getProductInfo() ); // OpenAI流式接口调用 Flux<String> stream = completionsClient.stream(prompt); StringBuilder orderInfo = new StringBuilder(); stream.subscribe( chunk -> orderInfo.append(chunk), e -> System.err.println("下单生成异常：" + e.getMessage()) ); // 阻塞等待流式输出（测试用，生产建议异步） try { Thread.sleep(2000); } catch (InterruptedException e) { Thread.currentThread().interrupt(); } context.setOrderInfo(orderInfo.toString().trim()); return context; } } |

（2）支付智能体（PaymentAgentService.java）

|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.service.agent; import com.alipay.api.AlipayClient; import com.alipay.api.request.AlipayTradeQueryRequest; import com.alipay.api.response.AlipayTradeQueryResponse; import com.shop.multiagent.dto.ShopAgentContext; import org.springframework.stereotype.Service; import java.util.HashMap; import java.util.Map; @Service public class PaymentAgentService { private final AlipayClient alipayClient; public PaymentAgentService(AlipayClient alipayClient) { this.alipayClient = alipayClient; } // 支付节点：调用支付宝MCP校验支付状态 public ShopAgentContext execute(ShopAgentContext context) { try { AlipayTradeQueryRequest request = new AlipayTradeQueryRequest(); Map<String, String> bizContent = new HashMap<>(); bizContent.put("out_trade_no", context.getOrderId()); bizContent.put("trade_no", context.getTradeNo()); request.setBizContent(bizContent.toString()); // 调用支付宝MCP真实接口 AlipayTradeQueryResponse response = alipayClient.execute(request); context.setPaymentSuccess(response.isSuccess() && "TRADE_SUCCESS".equals(response.getTradeStatus())); } catch (Exception e) { context.setPaymentSuccess(false); System.err.println("支付校验异常：" + e.getMessage()); } return context; } } |

（3）物流智能体（LogisticsAgentService.java）

|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.service.agent; import com.shop.multiagent.dto.ShopAgentContext; import org.springframework.ai.openai.OpenAiChatClient; import org.springframework.stereotype.Service; @Service public class LogisticsAgentService { private final OpenAiChatClient chatClient; public LogisticsAgentService(OpenAiChatClient chatClient) { this.chatClient = chatClient; } // 物流节点：调用支付宝MCP查物流+OpenAI生成结果 public ShopAgentContext execute(ShopAgentContext context) { if (!context.isPaymentSuccess()) { context.setLogisticsInfo("支付失败，暂不查询物流"); return context; } // 模拟调用支付宝物流查询MCP接口（真实场景替换为支付宝API） String alipayLogistics = String.format("订单%s物流状态：已揽收，快递公司：顺丰，运单号：SF123456789", context.getOrderId()); // OpenAI优化物流信息输出 String prompt = String.format("将以下物流信息整理为用户友好的格式：%s", alipayLogistics); context.setLogisticsInfo(chatClient.call(prompt)); return context; } } |

（4）交付智能体（DeliveryAgentService.java）

|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.service.agent; import com.shop.multiagent.dto.ShopAgentContext; import org.springframework.ai.openai.OpenAiCompletionsClient; import org.springframework.stereotype.Service; import reactor.core.publisher.Flux; @Service public class DeliveryAgentService { private final OpenAiCompletionsClient completionsClient; public DeliveryAgentService(OpenAiCompletionsClient completionsClient) { this.completionsClient = completionsClient; } // 交付节点：流式Prompt生成交付信息 public ShopAgentContext execute(ShopAgentContext context) { if (!context.isPaymentSuccess()) { context.setDeliveryInfo("支付失败，无法交付"); return context; } String prompt = String.format( "基于订单%s的物流信息[%s]，生成交付预估信息，包含：预计送达时间、签收提醒、异常处理方式", context.getOrderId(), context.getLogisticsInfo() ); // OpenAI流式接口调用 Flux<String> stream = completionsClient.stream(prompt); StringBuilder deliveryInfo = new StringBuilder(); stream.subscribe( chunk -> deliveryInfo.append(chunk), e -> System.err.println("交付生成异常：" + e.getMessage()) ); try { Thread.sleep(2000); } catch (InterruptedException e) { Thread.currentThread().interrupt(); } context.setDeliveryInfo(deliveryInfo.toString().trim()); return context; } } |

（5）客服智能体（CustomerServiceAgent.java）

|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.service.agent; import com.shop.multiagent.dto.ShopAgentContext; import org.springframework.ai.openai.OpenAiChatClient; import org.springframework.ai.openai.OpenAiEmbeddingClient; import org.springframework.stereotype.Service; import org.springframework.web.client.RestTemplate; import java.util.HashMap; import java.util.Map; @Service public class CustomerServiceAgent { private final OpenAiEmbeddingClient embeddingClient; private final OpenAiChatClient chatClient; private final RestTemplate restTemplate = new RestTemplate(); // RAG知识库API地址 private final String RAG_API_URL = "http://localhost:8080/api/rag/cs-knowledge"; public CustomerServiceAgent(OpenAiEmbeddingClient embeddingClient, OpenAiChatClient chatClient) { this.embeddingClient = embeddingClient; this.chatClient = chatClient; } // 客服节点：RAG知识库+OpenAI生成应答 public ShopAgentContext execute(ShopAgentContext context) { try { // 1. 生成上下文向量（RAG核心） String csQuestion = String.format("用户询问订单%s的交付与物流问题，订单信息：%s，物流：%s，交付：%s", context.getOrderId(), context.getOrderInfo(), context.getLogisticsInfo(), context.getDeliveryInfo()); float[] embedding = embeddingClient.embed(csQuestion).getEmbedding(); // 2. 调用RAG知识库API Map<String, Object> ragRequest = new HashMap<>(); ragRequest.put("orderId", context.getOrderId()); ragRequest.put("embedding", embedding); Map<String, String> ragContext = restTemplate.postForObject(RAG_API_URL, ragRequest, Map.class); // 3. 生成客服应答 String prompt = String.format("基于知识库：%s，生成友好的客服应答，回复用户关于订单%s的问题", ragContext.get("context"), context.getOrderId()); context.setCustomerServiceReply(chatClient.call(prompt)); } catch (Exception e) { context.setCustomerServiceReply("客服应答生成失败：" + e.getMessage()); } return context; } // 模拟RAG知识库API public Map<String, String> mockCsRagApi(Map<String, Object> request) { Map<String, String> response = new HashMap<>(); response.put("context", "1. 顺丰物流江浙沪1天达；2. 交付异常可联系400-123-4567；3. 签收后7天无理由退换"); return response; } } |

5. 流程编排（ShopWorkflowService.java）

|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.service.workflow; import com.shop.multiagent.dto.ShopAgentContext; import com.shop.multiagent.service.agent.*; import org.springframework.ai.stategraph.StateGraph; import org.springframework.stereotype.Service; @Service public class ShopWorkflowService { private final OrderAgentService orderAgent; private final PaymentAgentService paymentAgent; private final LogisticsAgentService logisticsAgent; private final DeliveryAgentService deliveryAgent; private final CustomerServiceAgent csAgent; public ShopWorkflowService(OrderAgentService orderAgent, PaymentAgentService paymentAgent, LogisticsAgentService logisticsAgent, DeliveryAgentService deliveryAgent, CustomerServiceAgent csAgent) { this.orderAgent = orderAgent; this.paymentAgent = paymentAgent; this.logisticsAgent = logisticsAgent; this.deliveryAgent = deliveryAgent; this.csAgent = csAgent; } // 构建StateGraph流程：下单→支付→物流→交付→客服 public StateGraph<ShopAgentContext> buildWorkflow() { return StateGraph.<ShopAgentContext>builder() // 注册5个节点 .step("order", context -> orderAgent.execute(context)) .step("payment", context -> paymentAgent.execute(context)) .step("logistics", context -> logisticsAgent.execute(context)) .step("delivery", context -> deliveryAgent.execute(context)) .step("customerService", context -> csAgent.execute(context)) // 定义执行顺序 .edge("order", "payment") .edge("payment", "logistics") .edge("logistics", "delivery") .edge("delivery", "customerService") // 起始节点 .start("order") .build(); } // 执行全流程 public ShopAgentContext runWorkflow(ShopAgentContext initialContext) { StateGraph<ShopAgentContext> graph = buildWorkflow(); return graph.execute(initialContext); } } |

6. 控制层（ShopAgentController.java）

|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent.controller; import com.shop.multiagent.dto.ShopAgentContext; import com.shop.multiagent.service.agent.CustomerServiceAgent; import com.shop.multiagent.service.workflow.ShopWorkflowService; import org.springframework.http.ResponseEntity; import org.springframework.web.bind.annotation.PostMapping; import org.springframework.web.bind.annotation.RequestBody; import org.springframework.web.bind.annotation.RequestMapping; import org.springframework.web.bind.annotation.RestController; import java.util.Map; @RestController @RequestMapping("/api/shop") public class ShopAgentController { private final ShopWorkflowService workflowService; private final CustomerServiceAgent csAgent; public ShopAgentController(ShopWorkflowService workflowService, CustomerServiceAgent csAgent) { this.workflowService = workflowService; this.csAgent = csAgent; } // 购物全流程接口 @PostMapping("/run-workflow") public ResponseEntity<ShopAgentContext> runWorkflow(@RequestBody ShopAgentContext context) { ShopAgentContext result = workflowService.runWorkflow(context); return ResponseEntity.ok(result); } // 客服RAG知识库模拟接口 @PostMapping("/rag/cs-knowledge") public ResponseEntity<Map<String, String>> csRagApi(@RequestBody Map<String, Object> request) { return ResponseEntity.ok(csAgent.mockCsRagApi(request)); } } |

7. 启动类（ShopMultiAgentApplication.java）

|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent; import org.springframework.boot.SpringApplication; import org.springframework.boot.autoconfigure.SpringBootApplication; @SpringBootApplication public class ShopMultiAgentApplication { public static void main(String[] args) { SpringApplication.run(ShopMultiAgentApplication.class, args); } } |

8. 测试类（ShopAgentFlowTest.java）

|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| package com.shop.multiagent; import com.shop.multiagent.dto.ShopAgentContext; import com.shop.multiagent.service.agent.*; import com.shop.multiagent.service.workflow.ShopWorkflowService; import org.junit.jupiter.api.Test; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.boot.test.context.SpringBootTest; @SpringBootTest public class ShopAgentFlowTest { @Autowired private OrderAgentService orderAgent; @Autowired private PaymentAgentService paymentAgent; @Autowired private LogisticsAgentService logisticsAgent; @Autowired private DeliveryAgentService deliveryAgent; @Autowired private CustomerServiceAgent csAgent; @Autowired private ShopWorkflowService workflowService; // 测试1：下单智能体 @Test public void testOrderAgent() { ShopAgentContext context = new ShopAgentContext(); context.setOrderId("SHOP_001"); context.setUserId("USER_1001"); context.setProductInfo("华为Mate60 Pro 128G"); ShopAgentContext result = orderAgent.execute(context); System.out.println("【下单节点】结果：" + result.getOrderInfo()); } // 测试2：支付智能体 @Test public void testPaymentAgent() { ShopAgentContext context = new ShopAgentContext(); context.setOrderId("SHOP_001"); context.setTradeNo("2024060122001410086000000001"); ShopAgentContext result = paymentAgent.execute(context); System.out.println("【支付节点】结果：" + result.isPaymentSuccess()); } // 测试3：物流智能体 @Test public void testLogisticsAgent() { ShopAgentContext context = new ShopAgentContext(); context.setOrderId("SHOP_001"); context.setPaymentSuccess(true); ShopAgentContext result = logisticsAgent.execute(context); System.out.println("【物流节点】结果：" + result.getLogisticsInfo()); } // 测试4：交付智能体 @Test public void testDeliveryAgent() { ShopAgentContext context = new ShopAgentContext(); context.setOrderId("SHOP_001"); context.setPaymentSuccess(true); context.setLogisticsInfo("订单SHOP_001物流状态：已揽收，快递公司：顺丰，运单号：SF123456789"); ShopAgentContext result = deliveryAgent.execute(context); System.out.println("【交付节点】结果：" + result.getDeliveryInfo()); } // 测试5：客服智能体 @Test public void testCsAgent() { ShopAgentContext context = new ShopAgentContext(); context.setOrderId("SHOP_001"); context.setOrderInfo("华为Mate60 Pro 128G，金额5999元"); context.setLogisticsInfo("顺丰揽收，运单号SF123456789"); context.setDeliveryInfo("预计6月3日送达"); ShopAgentContext result = csAgent.execute(context); System.out.println("【客服节点】结果：" + result.getCustomerServiceReply()); } // 测试6：全流程 @Test public void testFullWorkflow() { ShopAgentContext initialContext = new ShopAgentContext(); initialContext.setOrderId("SHOP_001"); initialContext.setUserId("USER_1001"); initialContext.setProductInfo("华为Mate60 Pro 128G"); initialContext.setTradeNo("2024060122001410086000000001"); ShopAgentContext result = workflowService.runWorkflow(initialContext); System.out.println("【全流程】下单结果：" + result.getOrderInfo()); System.out.println("【全流程】支付状态：" + result.isPaymentSuccess()); System.out.println("【全流程】物流信息：" + result.getLogisticsInfo()); System.out.println("【全流程】交付信息：" + result.getDeliveryInfo()); System.out.println("【全流程】客服应答：" + result.getCustomerServiceReply()); } } |

五、配置文件（application.yml）

|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| server: port: 8080 spring: ai: openai: api-key: {OPENAI_API_KEY} # 替换为真实Key alipay: app-id: {ALIPAY_APP_ID} # 替换为真实值 private-key: {ALIPAY_PRIVATE_KEY} public-key: {ALIPAY_PUBLIC_KEY} gateway-url: https://openapi.alipay.com/gateway.do |

六、核心总结

流程编排 ：基于 Spring AI StateGraph 实现 5 个智能体的顺序执行（下单→支付→物流→交付→客服），节点间通过 ShopAgentContext 传递数据，支持条件分支（如支付失败跳过后续节点）；

接口真实性：

下单 / 交付节点调用 OpenAI Completions 流式接口实现 Prompt 输出；

支付 / 物流节点调用支付宝 MCP 真实接口；

客服节点基于 OpenAI Embeddings + RAG 知识库实现智能应答；

分层设计：严格遵循 Controller→Service→Config 分层，SpringBoot 注解驱动开发，代码简洁易扩展；

测试覆盖：测试类覆盖单节点与全流程测试，确保每个智能体独立可用、协作流程顺畅。

七、运行说明

替换配置文件中的 OPENAI_API_KEY、支付宝密钥为真实值；

启动应用后，通过 POST /api/shop/run-workflow 调用全流程接口；

运行测试类可依次验证每个节点的执行效果，快速定位问题。

相关推荐
Fairy要carry2 小时前
面试-Prefill 和 Decode
人工智能
艾莉丝努力练剑2 小时前
文件描述符fd：跨进程共享机制
java·linux·运维·服务器·开发语言·c++
YaraMemo2 小时前
向量求导规则
人工智能·机器学习·矩阵·信号处理
工藤新一¹2 小时前
《操作系统》第一章(1)
java·服务器·前端
用户5191495848452 小时前
WordPress Top Store 主题高危漏洞利用工具 (CVE-2024-10673)
人工智能·aigc
琛説2 小时前
Web-Rooter：一种 IR + Lint 模式的 AI Agent 创新尝试【或许是下一个 AI 爆火方向】
前端·人工智能
nimadan122 小时前
**豆包seed写剧本2025指南，AI编剧工具实战应用解析**
人工智能·python
新缸中之脑2 小时前
分发：AI的终极护城河
人工智能
databook2 小时前
在AI的世界里，做一个真实的人
人工智能·程序员·创业
热门推荐
01GitHub 镜像站点02Qwen3.5 开源全解析：从 0.8B 到 397B，代际升级 + 全场景选型指南03OpenClaw 使用和管理 MCP 完全指南04本地部署 OpenClaw + DeepSeek-R1 完全指南05UV安装并设置国内源06OpenClaw macOS 完整安装与本地模型配置教程（实战版）07Openclow安装保姆级教程08OpenClaw Control UI安全上下文访问配置09在 Windows 上配置 Claude Code从安装到解决网络问题10OpenClaw 飞书机器人不回复消息？3 小时踩坑总结