Spring AI MCP using Streamable-HTTP

The Streamable HTTP transport enables MCP servers to run as independent processes that support multiple client connections via HTTP POST and GET requests. It also provides optional Server-Sent Events (SSE) for streaming multiple server messages, serving as a replacement for the older SSE transport.

Introduced in spec version 2025–03–26, this transport is particularly well-suited for applications that need to push dynamic updates about tools, resources, or prompts to connected clients.

MCP Server

pom.xml

<properties>
        <java.version>25</java.version>
        <spring-ai.version>1.1.0-M2</spring-ai.version>
</properties>

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-starter-mcp-server-webmvc</artifactId>
</dependency>

<dependencyManagement>
        <dependencies>
            <dependency>
                <groupId>org.springframework.ai</groupId>
                <artifactId>spring-ai-bom</artifactId>
                <version>${spring-ai.version}</version>
                <type>pom</type>
                <scope>import</scope>
            </dependency>
        </dependencies>
    </dependencyManagement>

• Create tool: Here, i am creating a simple addition tool

@Component
public class CalculatorTools {

    @McpTool(name = "add", description = "Add two numbers together")
    public int add(
            @McpToolParam(description = "First number", required = true) int a,
            @McpToolParam(description = "Second number", required = true) int b) {
        System.out.println("Adding " + a + " and " + b);
        return a + b;
    }
}

Update application.properties

spring.application.name=mcp-server
server.port=9091

spring.ai.mcp.server.type=sync
spring.ai.mcp.server.annotation-scanner.enabled=true
spring.ai.mcp.server.protocol=streamable
spring.ai.mcp.server.streamable-http.keep-alive-interval=30s

MCP Client

An MCP client with the Streamable API is basically a client application (like an IDE, web app, or agent) that connects to an MCP server using the Streamable HTTP transport.

Here’s what it means in practice:

• Client role → The MCP client is the consumer. It makes HTTP POST and GET requests to the MCP server to invoke tools, fetch resources, or interact with prompts.
• Streamable API → Instead of waiting for a single response per request, the client can receive multiple messages over time from the server. This is possible because of Server-Sent Events (SSE) support in the Streamable transport.
• Why it matters → The client isn’t limited to one-off request/response cycles. The server can push updates dynamically (e.g., tool status changes, resource updates, or new prompts), and the client can handle them in near real time.
• Example use case:
• An MCP client in a developer IDE subscribes to a code analysis tool.
• The MCP server sends incremental analysis results via SSE.
• The client UI updates continuously without polling

So, an MCP client with the Streamable API is just a client that speaks HTTP + SSE with an MCP server and can handle ongoing streams of server events, not just single-shot responses.

<properties>
    <java.version>25</java.version>
    <spring-ai.version>1.1.0-M2</spring-ai.version>
</properties>

 <dependency>
            <groupId>org.springframework.ai</groupId>
            <artifactId>spring-ai-starter-mcp-client</artifactId>
 </dependency>

<dependency>
            <groupId>org.springframework.ai</groupId>
            <artifactId>spring-ai-starter-model-openai</artifactId>
 </dependency>

 <dependencyManagement>
        <dependencies>
            <dependency>
                <groupId>org.springframework.ai</groupId>
                <artifactId>spring-ai-bom</artifactId>
                <version>${spring-ai.version}</version>
                <type>pom</type>
                <scope>import</scope>
            </dependency>
        </dependencies>
    </dependencyManagement>

Update application.properties

spring.application.name=mcp-client
server.port=9092

spring.ai.mcp.client.streamable-http.connections.server1.url=http://localhost:9091
spring.ai.mcp.client.streamable-http.connections.server1.endpoint=/mcp

spring.ai.openai.api-key=

Use ChatClient with openai to call mcp server add tool

@Bean
CommandLineRunner commandLineRunner( SyncMcpToolCallbackProvider toolCallbackProvider, ChatClient.Builder chatClientBuilder) {
    return args -> {
        ToolCallback[] toolCallbacks = toolCallbackProvider.getToolCallbacks();
        ChatClient chatClient = chatClientBuilder.defaultToolCallbacks(toolCallbacks)
                .build();

        String response = chatClient.prompt("Using add tool, calculate 2 + 2 ?")
                .call()
                .content();
        System.out.println(response);
    };
}

Github link: https://github.com/siddharth223437/spring-ai-example-apps