Java Trello MCP

A Spring Boot application that provides Trello integration as a Model Composition Platform (MCP) tool for Spring AI.

Features

• Create and manage Trello cards, lists, and checklists
• Fully integrated with Spring AI's Model Composition Platform
• RESTful API with Swagger UI documentation
• Configurable Trello API credentials

Prerequisites

• Java 24
• Maven 3.8+
• Trello account with API access

Configuration

Configure your Trello API credentials using one of these methods:

Environment Variables

export TRELLO_API_KEY=your_api_key
export TRELLO_TOKEN=your_token
export TRELLO_ORGANIZATION_ID=your_organization_id
export TRELLO_BOARD_ID=your_board_id
export TRELLO_BASE_URL=https://api.trello.com/1

Application Properties

Edit src/main/resources/application.yml:

trello_config:
  api_key: your_api_key
  token: your_token
  organization_id: your_organization_id
  board_id: your_board_id
  base_url: https://api.trello.com/1

Running the MCP

1. Build the application:

   ./mvnw clean package
   

   Copy

2. Run the application:

   java -jar target/java_trello_mcp-0.0.1-SNAPSHOT.jar
   

   Copy

   Or with Maven:

   ./mvnw spring-boot:run
   

   Copy

3. The MCP server will start on port 8080

API Documentation

Access Swagger UI documentation at:

http://localhost:8080/swagger-ui.html

Available MCP Tools

The following Trello tools are available for Spring AI:

• create_card - Create a new Trello card in a specific list
• create_list - Create a new list on a Trello board
• get_lists - Get all lists from a Trello board
• get_cards - Get all cards from a specific list
• get_checklists - Get all checklists from a specific card
• create_checklist - Create a new checklist for a card
• create_check_item - Create a new item in a checklist
• delete_card - Delete a specific card
• delete_checklist - Delete a specific checklist
• delete_check_item - Delete a specific checklist item
• archive_list - Archive a specific list

MCP Client Configuration

To configure an AI platform (like Claude, Cursor, or others) to use this MCP server, add the following configuration to your MCP client setup:

{
  "mcpServers": {
    "mcp-trello": {
      "command": "java",
      "args": [
        "-jar", "C:\\Users\\your_username\\path\\to\\java_trello_mcp-0.0.1-SNAPSHOT.jar"
      ],
      "env": {
        "TRELLO_API_KEY": "your_api_key",
        "TRELLO_TOKEN": "your_token",
        "TRELLO_ORGANIZATION_ID": "your_org_id",
        "TRELLO_BOARD_ID": "your_board_id",
        "TRELLO_BASE_URL": "https://api.trello.com/1"
      }
    }
  }
}

Make sure to:

1. Update the jar path to match your local environment
2. Replace the environment variables with your actual Trello credentials
3. Configure this in your AI platform's MCP settings