PluckierMCP

TonyKennah/PluckierMCP

3.1

If you are the rightful owner of PluckierMCP and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

An MCP server implementation using Spring Boot for horse race information and Google Cloud Storage interaction.

Build & Test Status

Pluckier MCP Server

This is a Spring Boot application that provides information about horse races. It uses Spring AI to expose tool functions for an AI agents and has REST endpoints to retrieve race information. It pulls daily racing data in JSON format stored in a GCS bucket.

Technology Stack

Simple enough code but it's all about the information, right?

  • Spring Framework
  • Java 17
  • Spring Boot 3.x
  • Spring AI
  • Google Cloud Storage
  • Maven

Prerequisites

  • Java Development Kit (JDK) 17 or later.
  • Apache Maven.
  • ⚠️ Access to the specific GCP Cloud storage location, with access credentials stored in the environment: GOOGLE_APPLICATION_CREDENTIALS

Building & Running the Project

You can build the project using the Maven wrapper:

./mvn clean install

Running the Application

Generally you run the server via an AI agent such as Gemini Cli, Claude Desktop, or ChatGPT. Via a Java command, example for Gemini (settings.json file).

{
    "mcpServers": {
        "pluckier": {
            "command": "java",
            "args": [
	            "-Dspring.ai.mcp.server.stdio=true",
                "-jar",
                "<PATH_TO>target\\mcp-server-0.0.1-SNAPSHOT.jar"
            ]
        }
    }
}

To run the application manually, use the Spring Boot Maven plugin:

./mvn spring-boot:run

Or a java command:

./java -jar target/mcp-server-0.0.1-SNAPSHOT.jar

The server will start on http://localhost:8080 which shows links to the log viewer and exposes the parameterised REST endpoints.

image

Live Log Viewer

The application provides a real-time log viewer to monitor server activity. This is particularly useful for observing the AI agent's behavior and the results of the tool function calls as they happen.

  1. Ensure the application is running.
  2. Open your web browser and navigate to:
    http://localhost:8080/logs.html
    
  3. The page will automatically connect to the server's WebSocket endpoint and display log messages as they are generated.
image

Spring AI Tools

image

Examples:

Ex 1: image

Ex 2: image

Code Quality

image