ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;display: table;padding: 0px 1em;color: rgb(63, 63, 63);">MCP Server开发实践ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;font-size: 16px;letter-spacing: 0.1em;color: rgb(63, 63, 63);">上一篇文章我们介绍了MCP的基本原理MCP：大模型的"万能接口"革命，但是对于开发者来说更关心如何实现我们的自己的MCP Server，接下来我们将使用MCP提供的java sdk和spring-ai来实现一个MCP Server。ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;display: table;padding: 0px 0.2em;color: rgb(255, 255, 255);background: rgb(15, 76, 129);">构建Spring Boot服务ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;font-size: 16px;letter-spacing: 0.1em;color: rgb(63, 63, 63);">技术选型：ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;font-size: 16px;color: rgb(63, 63, 63);" class="list-paddingleft-1">ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;font-size: 16px;text-indent: -1em;display: block;margin: 0.2em 8px;color: rgb(63, 63, 63);">• Spring Boot 3.4.2，因为Spring AI支持的Spring Boot版本为3.4.xingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;font-size: 16px;font-style: normal;padding: 1em;border-radius: 6px;color: rgba(0, 0, 0, 0.5);background: rgb(247, 247, 247);">ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;font-size: 1em;display: block;letter-spacing: 0.1em;color: rgb(63, 63, 63);">Spring AI supports Spring Boot 3.4.x. When Spring Boot 3.5.x is released, we will support that as well.ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;font-size: 16px;color: rgb(63, 63, 63);" class="list-paddingleft-1">ingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;font-size: 16px;text-indent: -1em;display: block;margin: 0.2em 8px;color: rgb(63, 63, 63);">• JDK 17，Spring Boot3需要使用JDK 17• spring-ai-mcp-server-webmvc-spring-boot-starter 1.0.0-M6，Spring AI支持三种方式提供服务，这里采用webmvc- • Standard Input/Output (STDIO) - spring-ai-starter-mcp-server
- • Spring MVC (Server-Sent Events) - spring-ai-starter-mcp-server-webmvc
- • Spring WebFlux (Reactive SSE) - spring-ai-starter-mcp-server-webflux
• httpclient5，将现有服务的HTTP接口暴露为MCP Serverpom.xml文件示例：
<dependencyManagement>
<dependencies>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-dependencies</artifactId>
<version>3.4.2</version>
<type>pom</type>
<scope>import</scope>
</dependency>
</dependencies>
</dependencyManagement>
<dependencies>
<!-- MCP -->
<dependency>
<groupId>org.springframework.ai</groupId>
<artifactId>spring-ai-mcp-server-webmvc-spring-boot-starter</artifactId>
<version>1.0.0-M6</version>
</dependency>
<!-- Lombok -->
<dependency>
<groupId>org.projectlombok</groupId>
<artifactId>lombok</artifactId>
<optional>true</optional>
</dependency>
<!-- Apache HttpClient -->
<dependency>
<groupId>org.apache.httpcomponents.client5</groupId>
<artifactId>httpclient5</artifactId>
</dependency>
<dependency>
<groupId>org.apache.httpcomponents.core5</groupId>
<artifactId>httpcore5</artifactId>
</dependency>
<dependency>
<groupId>org.apache.httpcomponents.core5</groupId>
<artifactId>httpcore5-h2</artifactId>
</dependency>
<!-- Spring Boot Starter -->
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-web</artifactId>
</dependency>
</dependencies>
MCP配置spring:
application:
name:mcp-name
ai:
mcp:
server:
name:mcp-name
version:1.0.0
type:SYNC
sse-endpoint:/sse
- • Server实现，这里直接通过转发的方式将现有的HTTP接口暴露为MCP Server
/**
*@date2025/03/27 14:52
**/
@Service
publicclassSmdMcpService{
@Autowired
privateRestTemplate restTemplate;
@Value("${smd.service.url}")
privateString smdServiceUrl;
@Tool(name = "getSmdInfo", description = "获取表结构信息")
publicStringgetSmdInfo(@ToolParam(description = "业务系统")String businessSystem,
@ToolParam(description = "表名")Set<String> tableNames){
Map<String, Object> params =newHashMap<>();
params.put("businessSystem", businessSystem);
params.put("tableNames", tableNames);
ResponseEntity<String> response = restTemplate.postForEntity(
smdServiceUrl +"/mcp/api/getSmdInfo",
params,
String.class);
returnresponse.getBody();
}
@Tool(name = "getCRUDCode", description = "根据表名生成增删改查代码")
publicList<Map<String, Object>>getCRUDByTable(@ToolParam(description = "业务系统")String businessSystem,
@ToolParam(description = "表名")Set<String> tableNames,
@ToolParam(description = "模块名，非必填")String moduleName
){
Map<String, Object> params =newHashMap<>();
params.put("businessSystem", businessSystem);
params.put("tableNames", tableNames);
params.put("moduleName", moduleName);
params.put("author","smd-mcp");
HttpEntity<Map<String, Object>> httpEntity = newHttpEntity<>(params);
ResponseEntity<List<Map<String, Object>>> response = restTemplate.exchange(
smdServiceUrl +"/mcp/api/crud",
HttpMethod.POST,
httpEntity,
newParameterizedTypeReference<List<Map<String, Object>>>() {});
returnresponse.getBody();
}
}
/**
* MCP配置类
*
*@authorAI Assistant
*@date2024/03/21
*/
@Configuration
@Slf4j
publicclassMcpConfig{
@Bean
publicToolCallbackProvidersmdToolCallbackProvider(SmdMcpService smdMcpService, RulesMcpService rulesMcpService){
returnMethodToolCallbackProvider.builder()
.toolObjects(smdMcpService, rulesMcpService)
.build();
};
}
测试MCP使用支持MCP的客户端进行测试，客户端支持情况可查看：https://modelcontextprotocol.io/clients 这里使用Cursor进行测试： 配置mcp.json {
"mcpServers":{
"mcp-name":{
"url":"http://localhost:8089/sse",
"env":{
"API_KEY":"value"
}
}
}
}
配置后即可看到MCP Server提供的Tools，如下图所示：
 MCP思考MCP 还处于发展初期，现阶段更重要的是生态构建，基于统一标准下构筑的生态也会正向的促进整个领域的发展。 对于普通开发者我们可以直接使用已有的MCP工具平台：https://mcp.so/ 对于企业，我们可以通过代理的方式将已有HTTP接口暴露为MCP Server： - • 零侵入改造：无需修改原有HTTP服务代码即可获得MCP能力
- • 跨模型兼容：让原本不支持MCP的传统服务获得与大模型生态无缝对接的能力
- • 低成本投入：已有业务接入MCP生态的改造周期大幅缩短
后续考虑将其做成MCP代理服务，通过简单配置即可将已有业务转换为MCP Server，为AI智能体打开新世界的大门。 |
