A Model Context Protocol (MCP) server that connects Trello with AI assistants like Claude Desktop, GitHub Copilot Chat, and other MCP-compatible clients.

• 📋 List all your Trello boards
• 🔍 Read board contents (lists and cards)
• ➕ Create new cards
• 🔄 Move cards between lists
• 💬 Add comments to cards
• 🗃️ Archive cards
• 🔗 Access boards as MCP resources
• 🐳 Docker support for easy deployment

• Node.js 20+ installed (for local development)
• Docker (for containerized deployment)
• A Trello account
• Trello API credentials (API Key and Token)

1. Clone this repository:

git clone https://github.com/lioarce01/trello-mcp-server.git
cd trello-mcp-server

2. Install dependencies:

npm install

3. Build the TypeScript code:

npm run build

1. Clone this repository:

git clone https://github.com/lioarce01/trello-mcp-server.git
cd trello-mcp-server

2. Build the Docker image:

docker build -t trello-mcp-server .

1. Get your API Key:

   • Go to https://trello.com/app-key
   • Copy your API Key

2. Get your Token:

   • On the same page, click on "Token" link
   • Authorize the application and copy your Token

Add the server configuration to your Claude Desktop config file:

Windows: %APPDATA%\Claude\claude_desktop_config.json
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Linux: ~/.config/claude/claude_desktop_config.json

{
  "mcp": {
    "servers": {
      "trello-mcp": {
        "command": "node",
        "args": ["absolute/path/to/the/project/dist/index.js"],
        "env": {
          "TRELLO_API_KEY": "your_api_key",
          "TRELLO_TOKEN": "your_token",
          "TRELLO_BASE_URL": "https://api.trello.com/1"
        }
      }
    }
  }
}

For Docker deployment, add this configuration:

{
  "mcp": {
    "servers": {
      "trello-mcp": {
        "command": "docker",
        "args": [
          "run",
          "--rm",
          "-i",
          "-e",
          "TRELLO_API_KEY=your_api_key",
          "-e",
          "TRELLO_TOKEN=your_token",
          "-e",
          "TRELLO_BASE_URL=https://api.trello.com/1",
          "trello-mcp-server"
        ]
      }
    }
  }
}

Add to your VS Code settings.json:

{
  "mcp": {
    "servers": {
      "trello-mcp": {
        "command": "node",
        "args": ["absolute/path/to/the/project/dist/index.js"],
        "env": {
          "TRELLO_API_KEY": "your_api_key",
          "TRELLO_TOKEN": "your_token",
          "TRELLO_BASE_URL": "https://api.trello.com/1"
        }
      }
    }
  }
}

{
  "mcp": {
    "servers": {
      "trello-mcp": {
        "command": "docker",
        "args": [
          "run",
          "--rm",
          "-i",
          "-e",
          "TRELLO_API_KEY=your_api_key",
          "-e",
          "TRELLO_TOKEN=your_token",
          "-e",
          "TRELLO_BASE_URL=https://api.trello.com/1",
          "trello-mcp-server"
        ]
      }
    }
  }
}

Important:

• Replace absolute/path/to/the/project/dist/index.js with the actual absolute path to your compiled server file (local installation)
• Replace YOUR_TRELLO_API_KEY and YOUR_TRELLO_TOKEN with your actual Trello credentials

# Build the image
docker build -t trello-mcp-server .

# Run with API key and token as arguments
docker run --rm -i --env-file .env trello-mcp-server

To test if your server works correctly:

1. Build the project:

npm run build

2. Run with credentials:

node dist/index.js

1. Build and run with Docker:

docker build -t trello-mcp-server .
docker run --rm -i --env-file .env trello-mcp-server

2. You should see:

MCP server connected and ready.

Note: The server will wait for MCP client connections. To exit, press Ctrl+C.

Once configured, you can interact with your Trello boards through natural language:

Show me all my Trello boards

What cards are in my "Project Management" board?

Create a new card called "Review documentation" in the "To Do" list

Move the "Bug fix" card to the "In Progress" list

Add a comment to the card saying "This needs urgent attention"

Archive the completed card "Setup database"

The server exposes your Trello boards as MCP resources that can be read by AI assistants:

• Resource URI: board:{boardId}
• Content: JSON containing all lists and cards for the board

npm run build

# Build Docker image
docker build -t trello-mcp-server .

To run the server directly (for testing):

# With npm
npm run build
node dist/index.js

# With pnpm
pnpm run build
node dist/index.js

# Run with docker (pass credentials as arguments)
docker run --rm -i --env-file .env trello-mcp-server

You can also create a development script in your package.json:

{
  "scripts": {
    "build": "tsc",
    "start": "node dist/index.js",
    "dev": "tsx ./src/index.ts",
    "docker:build": "docker build -t trello-mcp-server .",
    "docker:run": "docker run --rm -i --env-file .env trello-mcp-server"
  }
}

1. Check credentials: Make sure you're passing API Key and Token as arguments
2. Verify file path: Ensure the path in your MCP configuration is correct (local installation)
3. Build first: Always run npm run build or docker build before testing
4. Test standalone: Try running the server independently first
5. Restart client: Restart your MCP client (Claude Desktop/VS Code) after config changes

1. Image not found: Make sure you've built the Docker image first with docker build -t trello-mcp-server .
2. Arguments not passed: Ensure API key and token are passed as arguments after the image name
3. Permissions: Check that Docker has the necessary permissions to run containers

• Double-check your Trello API Key and Token
• Ensure the token has the necessary permissions
• Try regenerating your token if it's expired
• Verify credentials are properly passed as arguments

• Verify the board/card/list IDs are correct
• Check that you have write permissions to the Trello board
• Look at the console logs for detailed error messages

• Never commit your API credentials to version control
• Store credentials securely and rotate them regularly
• Use .env files and add them to .gitignore
• The server only requires the permissions you grant via the Trello token
• Consider using environment variables for credentials in production
• Docker containers run with non-root user for security

This project is licensed under the MIT License - see the LICENSE file for details.

Made with ❤️ for the MCP community