Location>code7788 >text

Comparison of experiences of MCP Server Java development frameworks (spring ai mcp and solon ai mcp)

Popularity：639 ℃/2025-04-21 11:53:59

Two currently known mcp-server java application development frameworks (ID class, the encapsulation experience is relatively simple):

  • spring-ai-mcp, supports java17 or above
  • Solon-ai-mcp, supports java8 or above (also supports integration into third-party frameworks such as springboot2, jfinal, etc.)

The following are two frameworks to build a weather query mcp tool service.

1. Spring ai mcp server (supports java17 or above)

Add key dependency packages (version number and springboot are independent)

<dependency>
   <groupId></groupId>
   <artifactId>spring-ai-mcp-server-spring-boot-starter</artifactId>
   <version>1.0.0-M6</version>
</dependency>

Add configuration (name the service endpoint)

: jdbc-mcp-server

Sample code (build the service and then publish it as ToolCallbackProvider)

@Service
 public class JdbcQueryService {
     @Tool(description = "Query weather forecast")
     public String getWeather(@ToolParam(description = "City Location") String location) {
         return "Sunny, 14 degrees";
     }
 }

 @Configuration
 public class McpConfig {
     @Bean
     ToolCallbackProvider jdbcQueryTools(JdbcQueryService jdbcQueryService) {
         return MethodToolCallbackProvider
                 .builder()
                 .toolObjects(jdbcQueryService)
                 .build();
     }
 }

2. Solon ai mcp server (supports java8 or above)

Add key dependency packages (version number is consistent with solon)

<dependency>
    <groupId></groupId>
    <artifactId>solon-ai-mcp</artifactId>
    <version>3.2.0</version>
</dependency>

Sample code (very similar to mvc development)

@McpServerEndpoint(name="mcp-case1", sseEndpoint = "/case1/sse")
 public class McpServerTool {
     @ToolMapping(description = "Query weather forecast")
     public String getWeather(@ToolParam(description = "City Location") String location) {
         return "Sunny, 14 degrees";
     }
 }

Solon ai mcp server supports multi-endpoints. It is just a service that can provide multiple sets of tools (for different scenarios, with better flexibility):

  • There can be a set of tools about weather
  • There can be another set of tools for maps

3. Summary

Comparison of development experience

Compare srping-ai-mcp solon-ai-mcp
Development Component-based development Component-based development
Configuration Configure via yaml Components, that is, configuration (can also refer to yaml configuration)
release Published as ToolCallbackProvider via configurator Components are released
jdk requirements jdk17 or above jdk8 or above
Endpoint Support It seems that there can only be one (within one service) Supports multiple endpoints (within one service)

The development of solon-ai-mcp is relatively simple and trinity. And supports multi-endpoints.

<Previous Next>