docs: enhance MCP server documentation with client usage examples

- Add comprehensive Boot Starter client usage examples
- Improve documentation clarity and organization
- Update configuration examples for both STDIO and SSE transports
- Standardize formatting and fix typos

Signed-off-by: Christian Tzolov <christian.tzolov@broadcom.com>
This commit is contained in:
Christian Tzolov
2025-02-10 14:59:55 +01:00
parent cf529d6940
commit fab7b1da33
2 changed files with 89 additions and 19 deletions

View File

@@ -4,17 +4,16 @@ A Spring Boot starter project demonstrating how to build a Model Context Protoco
For more information, see the [MCP Server Boot Starter](https://docs.spring.io/spring-ai/reference/api/mcp/mcp-server-boot-starter-docs.html) reference documentation.
## Prerequisites
- Java 17 or later
- Maven 3.6+
- Basic understanding of Spring Boot and Spring AI
- Maven 3.6 or later
- Understanding of Spring Boot and Spring AI concepts
- (Optional) Claude Desktop for AI assistant integration
## About Spring AI MCP Server Boot Starter
This project uses `spring-ai-mcp-server-spring-boot-starter`, which provides:
The `spring-ai-mcp-server-spring-boot-starter` provides:
- Automatic configuration of MCP server components
- Support for both synchronous and asynchronous operation modes
- STDIO transport layer implementation
@@ -40,7 +39,7 @@ src/
## Building and Running
The server uses STDIO transport mode and is typically started automatically by the client. You only need to build the server jar:
The server uses STDIO transport mode and is typically started automatically by the client. To build the server jar:
```bash
./mvnw clean install -DskipTests
@@ -53,7 +52,6 @@ The project demonstrates how to implement and register MCP tools using Spring's
```java
@Service
public class WeatherService {
@Tool(description = "Get weather forecast for a specific latitude/longitude")
public String getWeatherForecastByLocation(
double latitude, // Latitude coordinate
@@ -64,7 +62,7 @@ public class WeatherService {
@Tool(description = "Get weather alerts for a US state")
public String getAlerts(
String state // Two-letter US state code (e.g. CA, NY)
String state // Two-letter US state code (e.g., CA, NY)
) {
// Implementation
}
@@ -79,7 +77,7 @@ public class McpServerApplication {
}
```
The auto-configuration will automatically register these tools with the MCP server. You can have multiple beans producing lists of ToolCallbacks, and the auto-configuration will merge them.
The auto-configuration automatically registers these tools with the MCP server. You can have multiple beans producing lists of ToolCallbacks, and the auto-configuration will merge them.
## Available Tools
@@ -111,7 +109,7 @@ CallToolResult forecast = client.callTool(
```java
@Tool(description = "Get weather alerts for a US state")
public String getAlerts(
String state // Two-letter US state code (e.g. CA, NY)
String state // Two-letter US state code (e.g., CA, NY)
) {
// Returns active alerts including:
// - Event type
@@ -133,6 +131,8 @@ CallToolResult alerts = client.callTool(
### Java Client Example
#### Create MCP Client Manually
```java
// Create server parameters
ServerParameters stdioParams = ServerParameters.builder("java")
@@ -148,10 +148,26 @@ var transport = new StdioClientTransport(stdioParams);
var client = McpClient.sync(transport).build();
```
The [ClientStdio.java](src/test/java/org/springframework/ai/mcp/sample/client/ClientStdio.java) shows how to implement an MCP client manually.
The [ClientStdio.java](src/test/java/org/springframework/ai/mcp/sample/client/ClientStdio.java) demonstrates how to implement an MCP client manually.
For a better development experience, consider using the [MCP Client Boot Starters](https://docs.spring.io/spring-ai/reference/api/mcp/mcp-client-boot-starter-docs.html). These starters enable auto-configuration of multiple STDIO and/or SSE connections to MCP servers. See the [starter-default-client](../../client-starter/starter-default-client) and [starter-webflux-client](../../client-starter/starter-webflux-client) projects for examples.
#### Use MCP Client Boot Starter
Use the [starter-default-client](../../client-starter/starter-default-client) to connect to the weather `starter-stdio-server`:
1. Follow the `starter-default-client` readme instructions to build a `mcp-starter-default-client-0.0.1-SNAPSHOT.jar` client application.
2. Run the client using the configuration file:
```bash
java -Dspring.ai.mcp.client.stdio.connections.server1.command=java \
-Dspring.ai.mcp.client.stdio.connections.server1.args=-jar,/Users/christiantzolov/Dev/projects/spring-ai-examples/model-context-protocol/weather/starter-stdio-server/target/mcp-weather-stdio-server-0.0.1-SNAPSHOT.jar \
-Dai.user.input='What is the weather in NY?' \
-Dlogging.pattern.console= \
-jar mcp-starter-default-client-0.0.1-SNAPSHOT.jar
```
### Claude Desktop Integration
To integrate with Claude Desktop, add the following configuration to your Claude Desktop settings:
@@ -162,7 +178,7 @@ To integrate with Claude Desktop, add the following configuration to your Claude
"spring-ai-mcp-weather": {
"command": "java",
"args": [
"-Dspring.ai.mcp.server.transport=STDIO",
"-Dspring.ai.mcp.server.stdio=true",
"-Dspring.main.web-application-type=none",
"-Dlogging.pattern.console=",
"-jar",
@@ -175,13 +191,11 @@ To integrate with Claude Desktop, add the following configuration to your Claude
Replace `/absolute/path/to/` with the actual path to your built jar file.
## Configuration
### Application Properties
All properties are prefixed with `spring.ai.mcp.server`
All properties are prefixed with `spring.ai.mcp.server`:
```properties
# Required STDIO Configuration
@@ -205,12 +219,12 @@ logging.file.name=mcp-weather-stdio-server.log
### Key Configuration Notes
1. **STDIO Mode Requirements**:
1. **STDIO Mode Requirements**
- Disable web application type (`spring.main.web-application-type=none`)
- Disable Spring banner (`spring.main.banner-mode=off`)
- Clear console logging pattern (`logging.pattern.console=`)
2. **Server Type**:
2. **Server Type**
- `SYNC` (default): Uses `McpSyncServer` for straightforward request-response patterns
- `ASYNC`: Uses `McpAsyncServer` for non-blocking operations with Project Reactor support
@@ -221,4 +235,3 @@ logging.file.name=mcp-weather-stdio-server.log
- [MCP Client Boot Starter](https://docs.spring.io/spring-ai/reference/api/mcp/mcp-server-boot-client-docs.html)
- [Model Context Protocol Specification](https://modelcontextprotocol.github.io/specification/)
- [Spring Boot Auto-configuration](https://docs.spring.io/spring-boot/docs/current/reference/html/features.html#features.developing-auto-configuration)

View File

@@ -143,7 +143,9 @@ public class WeatherService {
You can connect to the weather server using either STDIO or SSE transport:
### WebFlux SSE Client
### Manual Clients
#### WebFlux SSE Client
For servers using SSE transport:
@@ -152,7 +154,7 @@ var transport = new WebFluxSseClientTransport(WebClient.builder().baseUrl("http:
var client = McpClient.sync(transport).build();
```
### STDIO Client
#### STDIO Client
For servers using STDIO transport:
@@ -177,6 +179,61 @@ The sample project includes example client implementations:
For a better development experience, consider using the [MCP Client Boot Starters](https://docs.spring.io/spring-ai/reference/api/mcp/mcp-client-boot-starter-docs.html). These starters enable auto-configuration of multiple STDIO and/or SSE connections to MCP servers. See the [starter-default-client](../../client-starter/starter-default-client) and [starter-webflux-client](../../client-starter/starter-webflux-client) projects for examples.
### Boot Starter Clients
Lets use the [starter-webflux-client](../../client-starter/starter-webflux-client) client to connect to our weather `starter-webflux-server`.
Follow the `starter-webflux-client` readme instruction to build a `mcp-starter-webflux-client-0.0.1-SNAPSHOT.jar` client application.
#### STDIO Transport
1. Create a `mcp-servers-config.json` configuration file with this content:
```json
{
"mcpServers": {
"weather-starter-webflux-server": {
"command": "java",
"args": [
"-Dspring.ai.mcp.server.stdio=true",
"-Dspring.main.web-application-type=none",
"-Dlogging.pattern.console=",
"-jar",
"/absolute/path/to/mcp-weather-starter-webflux-server-0.0.1-SNAPSHOT.jar"
]
}
}
}
```
2. Run the client using the configuration file:
```bash
java -Dspring.ai.mcp.client.stdio.servers-configuration=file:mcp-servers-config.json \
-Dai.user.input='What is the weather in NY?' \
-Dlogging.pattern.console= \
-jar mcp-starter-webflux-client-0.0.1-SNAPSHOT.jar
```
#### SSE (WebFlux) Transport
1. Start the `mcp-weather-starter-webflux-server`:
```bash
java -jar mcp-weather-starter-webflux-server-0.0.1-SNAPSHOT.jar
```
starts the MCP server on port 8080.
2. In another console start the client configured with SSE transport:
```bash
java -Dspring.ai.mcp.client.sse.connections.weather-server.url=http://localhost:8080 \
-Dlogging.pattern.console= \
-Dai.user.input='What is the weather in NY?' \
-jar mcp-starter-webflux-client-0.0.1-SNAPSHOT.jar
```
## Additional Resources
* [Spring AI Documentation](https://docs.spring.io/spring-ai/reference/)