trello

This Trello MCP server exposes your Trello boards as MCP resources and provides a set of tools to interact with Trello from AI assistants or MCP clients. You can list boards, inspect board contents (lists and cards), create and move cards, add comments, and archive cards or lists. Each operation is exposed via named tools such as list_boards, read_board, create_card, move_card, add_comment, and archive_card, allowing natural-language prompts to drive Trello activity. Resources are exposed as MCP URIs in the form board:{boardId}, which returns the JSON representation of a board's lists and cards for consumption by your AI assistant or client.

To use it, configure the MCP client (Claude Desktop, VS Code Copilot Chat, or other MCP-compatible tooling) to connect to the server, typically by referencing the node server and providing Trello credentials as environment variables or via an env-file. After starting the server, you can issue natural-language commands like: "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", or "Move the 'Bug fix' card to the 'In Progress' list". The available tools below map directly to Trello operations and are designed to be invoked by the MCP client based on your prompt.

Supported actions and corresponding tools include listing boards (list_boards), reading a board’s contents (read_board with boardId), creating lists (create_list), creating cards (create_card), moving cards (move_card), adding comments (add_comment), archiving cards (archive_card) and lists (archive_list), updating names (update_list_name, update_card_name), and deleting boards (delete_board). Resources propagate to AI assistants so you can treat boards as data sources and operate on them via natural language prompts.

Prerequisites:

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

Option A: Local Installation

1. Clone the repository and install dependencies:

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

2. Build the TypeScript code:

npm run build

3. Run the server locally:

node dist/index.js

Option B: Docker Installation

1. Clone the repository and build the Docker image:

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

2. Run the server in Docker (provide credentials via environment variables or an env file):

docker 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

Prerequisites recap:

• A Trello API Key and Token
• Either a Node.js environment or Docker for deployment
• Basic familiarity with editing configuration to supply credentials

Tips and common issues:

• Ensure Trello credentials are kept secure. Use environment variables or a secrets manager to avoid hard-coding keys.
• If using Docker, you can supply credentials via --env or an --env-file to keep sensitive data out of command history.
• The server exposes Trello boards as resources; you’ll access them in MCP clients using URIs like board:{boardId}.
• When updating configuration paths in Claude Desktop or VS Code Copilot Chat, point the node server to the built dist/index.js file and provide necessary environment variables for Trello access.
• If you encounter connection issues, verify that npm run build has completed successfully and that the server is listening for MCP client connections. Restart the MCP client after changing credentials or server configuration.
• For Docker troubleshooting, ensure the image is built before running and that environment variables are correctly passed to the container.