agentleFS
delorenjdelorenj/mcp-server-trello/skill/SKILL.md

mcp-server-trello

Skill446 starsChanged 3 months agoMITView on GitHub

Trello MCP Server skill for board discovery, card workflows, checklist management, comments, attachments, labels, members, board/workspace selection, and health monitoring through the bundled @delorenj/mcp-server-trello server.

  • Reads credentials

We split this skill into the rules it gives an agent

13 rules. About 0 are common, 3 are distinctive, and 10 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 10

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this skill when an agent needs to inspect or manage Trello boards through the bundled MCP server.

    line 8
  2. Distinctive:

    Before using Trello tools, verify that the local server has been installed.

    line 14
  3. Very specific:

    Check for a built server at: {XDG_DATA_HOME:-$HOME/.local/share}/mcp-server-trello-skill/server/build/index.js

    line 16
  4. Very specific:

    If it is missing, run: bash {skill-root}/scripts/install.sh

    line 18
  5. Very specific:

    Start with references/trello-mcp/README.md for routing.

    line 32
  6. Very specific:

    Use references/trello-mcp/configuration.md for authentication, install, and client setup.

    line 33
  7. Very specific:

    Use references/trello-mcp/api.md when you need tool names and parameters.

    line 35
  8. Very specific:

    Use references/trello-mcp/patterns.md for common board, card, checklist, and attachment workflows.

    line 36
  9. Very specific:

    Use references/trello-mcp/gotchas.md for date formats, IDs, rate limits, and recovery steps.

    line 38
  10. Distinctive:

    Always discover the board, list, card, checklist, label, or member ID through tools before writing changes.

    line 67
  11. Very specific:

    Prefer set_active_board after board discovery when several operations target the same board.

    line 69
  12. Very specific:

    Use start dates as YYYY-MM-DD.

    line 71
  13. Distinctive:

    Treat Trello API writes as external side effects.

    line 72

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Trello MCP Server Skill

Use this skill when an agent needs to inspect or manage Trello boards through the bundled MCP server. The skill carries its own server source in assets/source/ and installs it locally on first use.

Activation

Before using Trello tools, verify that the local server has been installed.

  1. Check for a built server at: {XDG_DATA_HOME:-$HOME/.local/share}/mcp-server-trello-skill/server/build/index.js
  2. If it is missing, run: bash {skill-root}/scripts/install.sh
  3. Confirm the MCP client has TRELLO_API_KEY and TRELLO_TOKEN configured. Optional values are TRELLO_BOARD_ID, TRELLO_WORKSPACE_ID, https_proxy, and HTTPS_PROXY.

The install script builds from assets/source/ when Bun is available. If Bun is not available, it falls back to the published package install path and creates a local wrapper at the same build/index.js check path.

Reading Order

Load only the reference needed for the current Trello task.

  • Start with references/trello-mcp/README.md for routing.
  • Use references/trello-mcp/configuration.md for authentication, install, and client setup.
  • Use references/trello-mcp/api.md when you need tool names and parameters.
  • Use references/trello-mcp/patterns.md for common board, card, checklist, and attachment workflows.
  • Use references/trello-mcp/gotchas.md for date formats, IDs, rate limits, and recovery steps.

Tool Routing

Choose tools by workflow, not by guessing IDs.

  • Board discovery: list_boards, set_active_board, get_active_board_info, list_workspaces, set_active_workspace, list_boards_in_workspace.
  • Lists and cards: get_lists, get_cards_by_list_id, get_card, add_card_to_list, update_card_details, move_card, archive_card, copy_card, add_cards_to_list, get_my_cards, get_card_history.
  • Checklists: create_checklist, get_checklist_items, add_checklist_item, find_checklist_items_by_description, get_acceptance_criteria, get_checklist_by_name, update_checklist_item, delete_checklist_item, copy_checklist.
  • Comments: add_comment, update_comment, delete_comment, get_card_comments.
  • Attachments: attach_image_to_card, attach_file_to_card, attach_image_data_to_card, download_attachment.
  • Labels and members: get_board_labels, create_label, update_label, delete_label, get_board_members, assign_member_to_card, remove_member_from_card.
  • Health: get_health, get_health_detailed, get_health_metadata, get_health_performance, perform_system_repair.

Agent Rules

  • Always discover the board, list, card, checklist, label, or member ID through tools before writing changes.
  • Prefer set_active_board after board discovery when several operations target the same board.
  • Use start dates as YYYY-MM-DD. Use due dates as full ISO 8601 timestamps.
  • Treat Trello API writes as external side effects. Confirm destructive actions such as archive, delete, or repair unless the user already authorized them.
  • Respect Trello rate limits: 300 requests per 10 seconds per API key and 100 requests per 10 seconds per token. The server queues requests, but agents should still batch discovery and avoid polling loops.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.