docs: Add MCP SDK documentation and architecture diagrams

Add comprehensive documentation for the MCP Java SDK implementation
Add architecture diagrams showing layered design and class relationships
Reorganize documentation structure with dedicated pages for:
Overview and introduction
MCP Java SDK reference
Spring AI MCP integration
Include detailed examples and usage patterns for both sync and async APIs
This commit is contained in:
Christian Tzolov 2025-01-03 00:25:41 +01:00
commit 48fad96fe4
6 changed files with 718 additions and 50 deletions

File diff suppressed because one or more lines are too long

After

Width:  |  Height:  |  Size: 81 KiB

File diff suppressed because one or more lines are too long

After

Width:  |  Height:  |  Size: 44 KiB

View file

@ -1,3 +1,5 @@
* xref:index.adoc[Overview]
* xref:overview.adoc[Overview]
* xref:mcp.adoc[MCP Java SDK]
* xref:spring-mcp.adoc[Spring AI MCP]
// ** xref:concepts.adoc[AI Concepts]
// * xref:getting-started.adoc[Getting Started]

View file

@ -0,0 +1,488 @@
= Java MCP SDK
Java SDK implementation of the Model Context Protocol, enabling seamless integration with language models and AI tools.
== Features
* Synchronous and Asynchronous MCP Client and MCP Server implementations
* Standard MCP operations support:
** Protocol version compatibility negotiation
** Tool discovery, execution, list change notifications
** Resource management with URI templates
** Roots list management and notifications
** Prompt handling and management
** Smampling support for AI model interactions
* Multiple transport implementations:
** Stdio-based transport for process-based communication
** SSE-based transport for HTTP streaming
=== Dependencies
Add the following dependency to your Maven project:
[source,xml]
----
<dependency>
<groupId>org.springframework.experimental</groupId>
<artifactId>mcp</artifactId>
<version>0.4.0-SNAPSHOT</version>
</dependency>
----
== Architecture
The SDK follows a layered architecture with clear separation of concerns:
image::MCP-layers.svg[width=400,align=center]
* *Transport Layer (McpTransport)*: Handles JSON-RPC message serialization/deserialization via StdioTransport (stdin/stdout) and SseTransport (HTTP streaming).
* *Session Layer (McpSession)*: Manages communication patterns and state using DefaultMcpSession implementation.
* *Client/Server Layer*: Both use McpSession for sync/async operations, with McpClient handling client-side protocol operations and McpServer managing server-side protocol operations.
Following class diagram illustrates the layered architecture of the MCP SDK, showing the relationships between core interfaces (McpTransport, McpSession), their implementations, and the client/server components. It highlights how the transport layer connects to sessions, which in turn support both synchronous and asynchronous client/server implementations.
image::spring-ai-mcp-uml-classdiagram.svg[width=800]
Key Interactions:
* *Client/Server Initialization*: Transport setup, protocol compatibility check, capability negotiation, and implementation details exchange.
* *Message Flow*: JSON-RPC message handling with validation, type-safe response processing, and error handling.
* *Resource Management*: Resource discovery, URI template-based access, subscription system, and content retrieval.
* *Prompt System*: Discovery, parameter-based retrieval, change notifications, and content management.
* *Tool Execution*: Discovery, parameter validation, timeout-aware execution, and result processing.
== MCP Client
The MCP Client is a key component in the Model Context Protocol (MCP) architecture, responsible for establishing and managing connections with MCP servers. It implements the client-side of the protocol, handling:
* Protocol version negotiation to ensure compatibility with servers
* Capability negotiation to determine available features
* Message transport and JSON-RPC communication
* Tool discovery and execution
* Resource access and management
* Prompt system interactions
* Optional features like roots management and sampling support
The client provides both synchronous and asynchronous APIs for flexibility in different application contexts.
[tabs]
======
Sync API::
+
[source,java]
----
// Create a sync client with custom configuration
McpSyncClient client = McpClient.using(transport)
.requestTimeout(Duration.ofSeconds(10))
.capabilities(ClientCapabilities.builder()
.roots(true) // Enable roots capability
.sampling() // Enable sampling capability
.build())
.sampling( request -> new CreateMessageResult(response))
.sync();
// Initialize connection
client.initialize();
// List available tools
ListToolsResult tools = client.listTools();
// Call a tool
CallToolResult result = client.callTool(
new CallToolRequest("calculator",
Map.of("operation", "add", "a", 2, "b", 3))
);
// List and read resources
ListResourcesResult resources = client.listResources();
ReadResourceResult resource = client.readResource(
new ReadResourceRequest("resource://uri")
);
// List and use prompts
ListPromptsResult prompts = client.listPrompts();
GetPromptResult prompt = client.getPrompt(
new GetPromptRequest("greeting", Map.of("name", "Spring"))
);
// Add/remove roots
client.addRoot(new Root("file:///path", "description"));
client.removeRoot("file:///path");
// Close client
client.closeGracefully();
----
Async API::
+
[source,java]
----
// Create an async client with custom configuration
McpAsyncClient client = McpClient.using(transport)
.requestTimeout(Duration.ofSeconds(10))
.capabilities(ClientCapabilities.builder()
.roots(true) // Enable roots capability
.sampling() // Enable sampling capability
.build())
.sampling( request -> new CreateMessageResult(response))
.toolsChangeConsumer(tools -> {
logger.info("Tools updated: {}", tools);
})
.resourcesChangeConsumer(resources -> {
logger.info("Resources updated: {}", resources);
})
.promptsChangeConsumer(prompts -> {
logger.info("Prompts updated: {}", prompts);
})
.async();
// Initialize connection
client.initialize()
.flatMap(initResult -> {
// List available tools
return client.listTools();
})
.flatMap(tools -> {
// Call a tool
return client.callTool(new CallToolRequest(
"calculator",
Map.of("operation", "add", "a", 2, "b", 3)
));
})
.flatMap(result -> {
// List and read resources
return client.listResources()
.flatMap(resources ->
client.readResource(new ReadResourceRequest("resource://uri"))
);
})
.flatMap(resource -> {
// List and use prompts
return client.listPrompts()
.flatMap(prompts ->
client.getPrompt(new GetPromptRequest(
"greeting",
Map.of("name", "Spring")
))
);
})
.flatMap(prompt -> {
// Add/remove roots
return client.addRoot(new Root("file:///path", "description"))
.then(client.removeRoot("file:///path"));
})
.doFinally(signalType -> {
// Close client
client.closeGracefully().subscribe();
})
.subscribe();
----
======
=== Client Transport
The transport layer handles the communication between MCP clients and servers, providing different implementations for various use cases. The client transport manages message serialization, connection establishment, and protocol-specific communication patterns.
[tabs]
======
STDIO::
+
[source,java]
----
// Create transport for process-based communication
ServerParameters params = ServerParameters.builder("npx")
.args("-y", "@modelcontextprotocol/server-everything", "dir")
.build();
McpTransport transport = new StdioClientTransport(params);
----
SSE::
+
[source,java]
----
// Create transport for HTTP streaming
WebClient.Builder webClientBuilder = WebClient.builder()
.baseUrl("http://your-mcp-server");
McpTransport transport = new SseClientTransport(webClientBuilder);
----
======
=== Client Capabilities
The client can be configured with various capabilities:
[source,java]
----
var capabilities = ClientCapabilities.builder()
.roots(true) // Enable filesystem roots support with list changes notifications
.sampling() // Enable LLM sampling support
.build();
----
==== Roots Support
Roots define the boundaries of where servers can operate within the filesystem:
[source,java]
----
// Add a root dynamically
client.addRoot(new Root("file:///path", "description"));
// Remove a root
client.removeRoot("file:///path");
// Notify server of roots changes
client.rootsListChangedNotification();
----
The roots capability allows servers to:
* Request the list of accessible filesystem roots
* Receive notifications when the root list changes
* Understand which directories and files they have access to
==== Sampling Support
Sampling enables servers to request LLM interactions ("completions" or "generations") through the client:
[source,java]
----
// Configure sampling handler
Function<CreateMessageRequest, CreateMessageResult> samplingHandler = request -> {
// Sampling implementation that interfaces with LLM
return new CreateMessageResult(response);
};
// Create client with sampling support
var client = McpClient.using(transport)
.capabilities(ClientCapabilities.builder()
.sampling()
.build())
.sampling(samplingHandler)
.build();
----
This capability allows:
* Servers to leverage AI capabilities without requiring API keys
* Clients to maintain control over model access and permissions
* Support for both text and image-based interactions
* Optional inclusion of MCP server context in prompts
== MCP Server
The MCP Server is a foundational component in the Model Context Protocol (MCP) architecture that provides tools, resources, and capabilities to clients. It implements the server-side of the protocol, responsible for:
* Exposing tools that clients can discover and execute
* Managing resources with URI-based access patterns
* Providing prompt templates and handling prompt requests
* Supporting capability negotiation with clients
* Implementing server-side protocol operations
* Managing concurrent client connections
* Providing structured logging and notifications
The server supports both synchronous and asynchronous APIs, allowing for flexible integration in different application contexts. It can expose various capabilities such as file system operations, AI model interactions, and custom tool implementations.
[tabs]
======
Sync API::
+
[source,java]
----
// Create a server with custom configuration
McpSyncServer syncServer = McpServer.using(transport)
.info("my-server", "1.0.0")
.capabilities(ServerCapabilities.builder()...build())
.tools(new CalculatorTool())
.resources(resourceRegistration)
.prompts(promptRegistration)
.sync();
// Add a tool handler at runtime
syncServer.addTool(new CalculatorTool());
// Remove a tool handler at runtime
syncServer.removeTool("calculator");
// Add a resource at runtime
syncServer.addResource(resourceRegistration);
// Remove a resource at runtime
syncServer.removeResource(resourceUri);
// Add a prompt at runtime
syncServer.addPrompt(promptRegistration);
// Remove a prompt at runtime
syncServer.removePrompt(promptName);
// Graceful shutdown
syncServer.closeGracefully();
----
Async API::
+
[source,java]
----
// Create an async server with custom configuration
McpAsyncServer asyncServer = McpServer.using(transport)
.info("my-server", "1.0.0")
.capabilities(ServerCapabilities.builder()...build())
.tools(new CalculatorTool())
.resources(resourceRegistration)
.prompts(promptRegistration)
.async();
// Add a tool handler at runtime
asyncServer.addTool(new CalculatorTool())
.doOnSuccess(v -> logger.info("Tool added"))
.subscribe();
// Remove a tool handler at runtime
asyncServer.removeTool("calculator")
.doOnSuccess(v -> logger.info("Tool removed"))
.subscribe();
// Add a resource at runtime
asyncServer.addResource(resourceRegistration)
.doOnSuccess(v -> logger.info("Resource added"))
.subscribe();
// Remove a resource at runtime
asyncServer.removeResource(resourceUri)
.doOnSuccess(v -> logger.info("Resource removed"))
.subscribe();
// Add a prompt at runtime
asyncServer.addPrompt(promptRegistration)
.doOnSuccess(v -> logger.info("Prompt added"))
.subscribe();
// Remove a prompt at runtime
asyncServer.removePrompt(promptName)
.doOnSuccess(v -> logger.info("Prompt removed"))
.subscribe();
// Notify clients of changes
asyncServer.notifyToolsListChanged().subscribe();
asyncServer.notifyResourcesListChanged().subscribe();
asyncServer.notifyPromptsListChanged().subscribe();
// Graceful shutdown
asyncServer.closeGracefully().subscribe();
----
======
=== Server Transport
The server transport layer implements the server-side communication protocols, enabling reliable message exchange with MCP clients. It provides implementations for different communication patterns while ensuring proper message handling, connection management, and protocol compliance.
[tabs]
======
STDIO::
+
[source,java]
----
// Create transport with custom ObjectMapper
ObjectMapper mapper = new ObjectMapper();
StdioServerTransport transport = new StdioServerTransport(mapper);
----
+
Provides bidirectional JSON-RPC message handling over standard input/output streams with non-blocking message processing, serialization/deserialization, and graceful shutdown support.
SSE::
+
[source,java]
----
// Create SSE transport
ObjectMapper mapper = new ObjectMapper();
String messageEndpoint = "/mcp/message";
SseServerTransport transport = new SseServerTransport(mapper, messageEndpoint);
// Get router function for web server configuration
RouterFunction<?> router = transport.getRouterFunction();
----
+
Implements the MCP HTTP with SSE transport specification, providing concurrent client connections through SSE endpoints with message routing, session management, and graceful shutdown capabilities.
======
=== Server Capabilities
The server can be configured with various capabilities:
[source,java]
----
var capabilities = ServerCapabilities.builder()
.resources(false, true) // Resource support with list changes notifications
.tools(true) // Tool support with list changes notifications
.prompts(true) // Prompt support with list changes notifications
.logging() // Enable logging support (enabled by default with loging level INFO)
.build();
----
==== Logging Support
The server provides structured logging capabilities that allow sending log messages to clients with different severity levels:
[source,java]
----
// Send a log message to clients
server.loggingNotification(LoggingMessageNotification.builder()
.level(LoggingLevel.INFO)
.logger("custom-logger")
.data("Custom log message")
.build());
----
Clients can control the minimum logging level they receive through the `mcpClient.setLoggingLevel(level)` request. Messages below the set level will be filtered out.
Supported logging levels (in order of increasing severity): DEBUG (0), INFO (1), NOTICE (2), WARNING (3), ERROR (4), CRITICAL (5), ALERT (6), EMERGENCY (7)
==== Tool Registration
[source,java]
----
var toolRegistration = new ToolRegistration(
new Tool("calculator", "Basic calculator", Map.of(
"operation", "string",
"a", "number",
"b", "number"
)),
arguments -> {
// Tool implementation
return new CallToolResult(result, false);
}
);
----
==== Resource Registration
[source,java]
----
var resourceRegistration = new ResourceRegistration(
new Resource("custom://resource", "name", "description", "mime-type", null),
request -> {
// Resource read implementation
return new ReadResourceResult(contents);
}
);
----
==== Prompt Registration
[source,java]
----
var promptRegistration = new PromptRegistration(
new Prompt("greeting", "description", List.of(
new PromptArgument("name", "description", true)
)),
request -> {
// Prompt implementation
return new GetPromptResult(description, messages);
}
);
----
== Error Handling
The SDK provides comprehensive error handling through the McpError class, covering protocol compatibility, transport communication, JSON-RPC messaging, tool execution, resource management, prompt handling, timeouts, and connection issues. This unified error handling approach ensures consistent and reliable error management across both synchronous and asynchronous operations.

View file

@ -0,0 +1,128 @@
= Java & Spring MCP
Java SDK for the Model Context Protocol (MCP), providing seamless integration between Java and Spring applications and MCP-compliant AI resources and tools.
== Overview
Provides Java and Spring Framework integration for the link:https://modelcontextprotocol.org/docs/concepts/architecture[Model Context Protocol]. It enables Java applications to interact with AI models and tools through a standardized interface, supporting both synchronous and asynchronous communication patterns.
image::spring-ai-mcp-architecture.jpg[Spring AI MCP Architecture,600]
== Modules
The project consists of two main modules:
=== xref:mcp.adoc[MCP Java SDK]
Java implementation of the Model Context Protocol specification. It includes:
* Synchronous and asynchronous link:https://github.com/spring-projects-experimental/spring-ai-mcp/blob/main/mcp/README.md#client-usage-examples[MCP Client] and link:https://github.com/spring-projects-experimental/spring-ai-mcp/blob/main/mcp/README.md#server-usage-examples[MCP Server] implementations
* Standard MCP operations support (tool discovery, resource management, prompt handling, structured logging). Support for request and notification handling.
* link:https://spec.modelcontextprotocol.io/specification/basic/transports/#stdio[Stdio] and link:https://spec.modelcontextprotocol.io/specification/basic/transports/#http-with-sse[SSE] transport implementations.
=== xref:spring-mcp.adoc[Spring AI MCP]
The Spring integration module provides Spring-specific functionality:
* Integration with Spring AI's function calling system
* Spring-friendly abstractions for MCP clients
* Auto-configurations (WIP)
== Installation
Add the following dependencies to your Maven project:
[tabs]
======
Maven::
+
[source,xml]
----
<!-- For core MCP functionality -->
<dependency>
<groupId>org.springframework.experimental</groupId>
<artifactId>mcp</artifactId>
<version>0.4.0-SNAPSHOT</version>
</dependency>
----
+
[source,xml]
----
<!-- For Spring AI integration -->
<dependency>
<groupId>org.springframework.experimental</groupId>
<artifactId>spring-ai-mcp</artifactId>
<version>0.4.0-SNAPSHOT</version>
</dependency>
----
+
This is a milestone release. It is not available on Maven Central. Add this repository to your POM:
+
[source,xml]
----
<repositories>
<repository>
<id>spring-milestones</id>
<name>Spring Milestones</name>
<url>https://repo.spring.io/libs-milestone-local</url>
<snapshots>
<enabled>false</enabled>
</snapshots>
</repository>
</repositories>
----
Gradle::
+
[source,groovy]
----
dependencies {
implementation 'org.springframework.experimental:mcp'
}
----
Or for Spring AI integration:
+
[source,groovy]
----
dependencies {
implementation 'org.springframework.experimental:spring-ai-mcp'
}
----
======
== Examples
Explore these MCP examples in the link:https://github.com/spring-projects/spring-ai-examples/tree/main/model-context-protocol[spring-ai-examples/model-context-protocol] repository:
* link:https://github.com/spring-projects/spring-ai-examples/tree/main/model-context-protocol/sqlite/simple[SQLite Simple] - Demonstrates LLM integration with a database
* link:https://github.com/spring-projects/spring-ai-examples/tree/main/model-context-protocol/sqlite/chatbot[SQLite Chatbot] - Interactive chatbot with SQLite database interaction
* link:https://github.com/spring-projects/spring-ai-examples/tree/main/model-context-protocol/filesystem[Filesystem] - Enables LLM interaction with local filesystem folders and files
* link:https://github.com/spring-projects/spring-ai-examples/tree/main/model-context-protocol/brave[Brave] - Enables natural language interactions with Brave Search, allowing you to perform internet searches.
* link:./spring-ai-mcp-sample/[Spring-ai-mcp-sample] - Showcases how to create and use MCP servers and clients with different transport modes and capabilities.
== Documentation
* link:mcp/README.md[Java MCP SDK documentation]
** link:docs/ref-index.md[Reference documentation]
* link:spring-ai-mcp/README.md[Spring Integration documentation]
== Development
=== Building from Source
[source,bash]
----
mvn clean install
----
=== Running Tests
[source,bash]
----
mvn test
----
== License
This project is licensed under the link:LICENSE[Apache License 2.0].

View file

@ -0,0 +1,41 @@
= Spring AI MCP Spring
:page-title: Spring AI MCP Spring
:doctype: book
:icons: font
:source-highlighter: highlight.js
:toc: left
Spring Integration module for Model Control Protocol (MCP) that provides Spring-specific functionality for working with MCP clients.
== Overview
The `spring-ai-mcp` module is part of the https://github.com/spring-projects-experimental/spring-ai-mcp[Spring AI MCP] project. It provides Spring Framework integration for the Model Control Protocol (MCP), enabling seamless integration of MCP functionality within Spring applications.
== Features
* Spring integration for MCP clients
* Function callback implementation for tool invocations
* Seamless integration with Spring AI's function calling capabilities
== Main Components
=== McpFunctionCallback
The `McpFunctionCallback` class is the primary component of this module. It implements Spring AI's `FunctionCallback` interface and provides:
* Integration between Spring AI's function calling system and MCP tools
* Automatic conversion between JSON and Java objects for tool arguments
* Synchronous tool execution support
== Usage
To use this module, add the following dependency to your Maven project:
[source,xml]
----
<dependency>
<groupId>org.springframework.experimental</groupId>
<artifactId>spring-ai-mcp</artifactId>
<version>0.4.0-SNAPSHOT</version>
</dependency>
----