MCP Server Trello
MCP 服务器 Trello
一个模型上下文协议 (MCP) 服务器,提供与 Trello 看板交互的工具。该服务器能够与 Trello 的 API 无缝集成,同时自动处理速率限制、类型安全和错误处理。
特征
完整的 Trello Board 集成:与卡片、列表和板活动进行交互
内置速率限制:遵守 Trello 的 API 限制(每个 API 密钥 300 个请求/10 秒,每个令牌 100 个请求/10 秒)
类型安全实现:用 TypeScript 编写,具有全面的类型定义
输入验证:对所有 API 输入进行强大的验证
错误处理:优雅的错误处理,提供信息丰富的消息
Related MCP server: MCP Trello
安装
npm install @modelcontextprotocol/mcp-server-trello配置
使用以下配置将服务器添加到您的 MCP 设置文件中:
{
"mcpServers": {
"trello": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-trello"],
"env": {
"TRELLO_API_KEY": "your-api-key",
"TRELLO_TOKEN": "your-token",
"TRELLO_BOARD_ID": "your-board-id"
}
}
}
}必需的环境变量
TRELLO_API_KEY:您的 Trello API 密钥(从https://trello.com/app-key获取)TRELLO_TOKEN:您的 Trello 令牌(使用您的 API 密钥生成)TRELLO_BOARD_ID:要交互的 Trello 板的 ID(可在板 URL 中找到)
可用工具
通过列表ID获取卡片
从特定列表中获取所有卡片。
{
name: 'get_cards_by_list_id',
arguments: {
listId: string // ID of the Trello list
}
}获取列表
从配置的板中检索所有列表。
{
name: 'get_lists',
arguments: {}
}获取最近的活动
获取论坛上的最近活动。
{
name: 'get_recent_activity',
arguments: {
limit?: number // Optional: Number of activities to fetch (default: 10)
}
}将卡片添加到列表
将新卡片添加到指定列表。
{
name: 'add_card_to_list',
arguments: {
listId: string, // ID of the list to add the card to
name: string, // Name of the card
description?: string, // Optional: Description of the card
dueDate?: string, // Optional: Due date (ISO 8601 format)
labels?: string[] // Optional: Array of label IDs
}
}更新卡片详情
更新现有卡的详细信息。
{
name: 'update_card_details',
arguments: {
cardId: string, // ID of the card to update
name?: string, // Optional: New name for the card
description?: string, // Optional: New description
dueDate?: string, // Optional: New due date (ISO 8601 format)
labels?: string[] // Optional: New array of label IDs
}
}存档卡
将卡片发送到档案馆。
{
name: 'archive_card',
arguments: {
cardId: string // ID of the card to archive
}
}将列表添加到板
向板上添加新列表。
{
name: 'add_list_to_board',
arguments: {
name: string // Name of the new list
}
}存档列表
将列表发送至档案馆。
{
name: 'archive_list',
arguments: {
listId: string // ID of the list to archive
}
}获取我的卡片
获取分配给当前用户的所有卡。
{
name: 'get_my_cards',
arguments: {}
}速率限制
服务器实现了令牌桶算法来进行速率限制,以符合 Trello 的 API 限制:
每个 API 密钥每 10 秒 300 个请求
每个令牌每 10 秒 100 个请求
速率限制是自动处理的,如果达到限制,请求将排队。
错误处理
服务器针对各种场景提供了详细的错误消息:
输入参数无效
超出速率限制
API 身份验证错误
网络问题
无效的看板/列表/卡片 ID
发展
先决条件
Node.js 16 或更高版本
npm 或 yarn
设置
克隆存储库
git clone https://github.com/modelcontextprotocol/server-trello.git
cd server-trello安装依赖项
npm install构建项目
npm run build运行测试
npm test贡献
欢迎贡献代码!请阅读我们的贡献指南,详细了解我们的行为准则以及提交 Pull Request 的流程。
执照
该项目根据 MIT 许可证获得许可 - 有关详细信息,请参阅LICENSE文件。
致谢
Available Tools
9 toolsadd_card_to_listC
Add a new card to a specified list
Inputs: listId (required), name (required), description, dueDate, +1 more.
add_list_to_boardC
Add a new list to the board
Inputs: name (required).
archive_cardC
Send a card to the archive
Inputs: cardId (required).
archive_listC
Send a list to the archive
Inputs: listId (required).
get_cards_by_list_idB
Fetch cards from a specific Trello list
Inputs: listId (required).
get_listsB
Retrieve all lists from the specified board
Inputs: no named parameters.
get_my_cardsA
Fetch all cards assigned to the current user
Inputs: no named parameters.
get_recent_activityC
Fetch recent activity on the Trello board
Inputs: limit.
update_card_detailsC
Update an existing card's details
Inputs: cardId (required), description, dueDate, labels, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 9 tools
Each tool has a clearly distinct purpose with no ambiguity: adding cards/lists, archiving cards/lists, fetching cards by list/board/user, getting activity, and updating cards. The descriptions reinforce this by specifying unique actions on different Trello entities.
The naming is mostly consistent with a verb_noun pattern (e.g., add_card_to_list, archive_card, get_cards_by_list_id), but there are minor deviations like 'get_my_cards' (which could be 'get_cards_by_user' for full consistency) and 'get_recent_activity' (which doesn't follow the verb_noun structure as tightly).
With 9 tools, this is well-scoped for a Trello server, covering core operations like CRUD for cards and lists, archiving, and activity tracking. Each tool earns its place without feeling excessive or insufficient for managing Trello boards.
The toolset provides strong coverage for Trello's domain, including create (add_card_to_list, add_list_to_board), read (get_cards_by_list_id, get_lists, get_my_cards, get_recent_activity), update (update_card_details), and delete/archive (archive_card, archive_list). A minor gap is the lack of a tool to update list details, but agents can work around this.
Related MCP Connectors
Plan trips with your AI: cards, itinerary, board, files and comments, with role-aware OAuth access.
Log time on Trello cards and read time reports. Free during early access.
Task management for people and AI agents, with scoped OAuth access to issues, projects, and docs.
Task management for people and AI agents, with scoped OAuth access to issues, projects, and docs.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceFacilitates interaction with Trello boards via the Trello API, offering features like rate limiting, type safety, input validation, and error handling for seamless management of cards, lists, and board activities.5,731 npm446MIT
- AlicenseNot gradedqualityDmaintenanceEnables interaction with Trello boards through comprehensive card, list, and board management tools. Includes built-in rate limiting, type safety, and support for operations like creating cards, updating details, managing members, and tracking board activity.213 npm10MIT
- AlicenseNot gradedqualityDmaintenanceEnables interaction with Trello boards, lists, and cards through the Trello REST API. Supports board management, card operations, member management, labels, and checklists through natural language.1MIT
- AlicenseNot gradedqualityDmaintenanceProvides seamless integration with Trello's API to manage boards, lists, and cards through natural language. It supports full CRUD operations, card movement, and the ability to load Trello resources directly into an LLM's context for analysis.MIT