Can AI agents automate Trello?

Trello is automatable today through its REST API at api.trello.com/1. Boards, lists, cards, actions, members, and webhooks are read/write under a user's token. Applications identify themselves with a public API key from a Power-Up and obtain a user token through the 1/authorize delegated flow or OAuth 1.0. Tokens should be kept secret; Trello documents query-parameter, Authorization-header, and PUT/POST-body delivery. Rate limits are 300 requests per 10 seconds per key and 100 per 10 seconds per token, with 429 responses when exceeded. Webhooks POST signed JSON when a watched model changes. This record is Trello only. Jira, Confluence, and Atlassian Rovo MCP are separate identities. Atlassian also hosts a first-party Trello MCP server at https://mcp.trello.com/v1. It uses OAuth 2.0 consent, is limited to one workspace per connection at launch, respects existing Trello permissions, and does not support destructive delete or API-token MCP auth. Current tools cover members, workspaces, boards, lists, cards, checklists, search, Inbox, and Planner; comments, attachments, custom fields, and activity history are documented as coming later. Independent third-party MCP servers also wrap the REST API. A first-party CLI, workflow-connector catalogs, and computer-use viability were not established.

Best path today: api · Overall automatability: 8/10