Spring-AI-showcase-with-simple

This MCP server exposes a small set of real-time date and time utilities (DateTimeTools) that a local LLM can call to obtain current temporal information. The tools include endpoints to fetch the current time, current date, current year, and the current weekday. The server is designed to act as an aggregation layer so your LLM can request real-time data without maintaining it locally. To use it, run the containerized server (or the recommended docker-compose setup) and connect your LM Studio or any MCP-capable client to the provided MCP endpoint. Once connected, the tools will appear as available capabilities that your LLM can invoke in prompts to enrich responses with fresh temporal data.

Prerequisites:

• Docker or Podman installed
• Optional: Docker Compose if you want to run the multi-service setup locally
• Basic knowledge of running MCP servers locally

Option A — Run with Docker (single container):

1. Pull and run the image directly (as per the repository’s suggested image): docker run -d --name spring-ai-showcase -p 58090:58090 schieblerchris/spring-ai-showcase-with-simple:latest
2. Verify the service is up by curling the MCP endpoint: curl http://localhost:58090/mcp
3. In your LM Studio or MCP client, register the server at http://localhost:58090/mcp using the mcp.json configuration provided in the repository.

Option B — Run with Docker Compose (recommended for local database and multi-service setup):

1. Navigate to the docker folder containing docker-compose.yml
2. Start services: docker-compose up -d
3. Access the MCP endpoint at http://localhost:58090/mcp (or as defined in the compose file) and ensure the tools are exposed to your LLM.

Option C — Local build (Java, if you prefer not to use Docker):

1. Ensure JDK 21+ and Maven installed
2. Build the project: mvn clean package -DskipTests
3. Run the Spring Boot application (adjust the path to the built jar as needed): java -jar target/spring-ai-showcase-with-simple-0.0.1-SNAPSHOT.jar
4. The MCP API should be available at the configured port (commonly http://localhost:58090/mcp).

Note: The README suggests using Docker Compose for local deployment, especially to host the accompanying database. Adjust ports and image names as needed for your environment.

Tips and considerations:

• If you encounter network issues with LM Studio, ensure the MCP endpoint is reachable from your network and that any firewalls allow the port used by the server.
• The tools exposed by this MCP server are limited to DateTimeTools (current time, date, year, weekday). If you need additional utilities, you may extend the server with more endpoints.
• In a multi-user or networked environment, consider configuring LM Studio’s mcp.json with the same endpoint you expose in your docker-compose setup.
• If using Docker, ensure the image tag matches the latest release or your built image version to avoid compatibility issues with Spring AI 1.x.
• When running behind a proxy, you may need to configure the proxy settings for Docker or Java accordingly.