ClickUp MCP Server
Integrates with ClickUp API to manage tasks, workspaces, spaces, folders, and lists, enabling AI assistants to perform discovery, read, and write operations across ClickUp hierarchy.
README
ClickUp MCP Server
A Model Context Protocol (MCP) server that integrates with the ClickUp API, enabling AI assistants to manage tasks across your entire ClickUp workspace hierarchy.
Features
Discovery Operations
- list_workspaces - List all workspaces/teams you have access to
- list_spaces - List all spaces in a workspace
- list_folders - List all folders in a space
- list_lists - List all lists in a space or folder
- search_tasks - Search tasks across a workspace with filters
Read Operations
- list_today_tasks - Retrieve all tasks due today (workspace-wide or list-specific)
- list_overdue_tasks - Retrieve all tasks past their due date
- list_no_due_date_tasks - Retrieve all tasks without a due date
Write Operations
- create_task - Create a new task with optional description, due date, priority, and assignee
- update_task - Update task status, due date, or priority
- add_comment - Add a comment to an existing task
Prerequisites
- Node.js v20 or higher
- A ClickUp account with API access
- A ClickUp API token
Installation
-
Clone this repository:
git clone <repository-url> cd clickup-mcp -
Install dependencies:
npm install -
Create a
.envfile from the example:cp .env.example .env -
Configure your environment variables in
.env:CLICKUP_API_TOKEN=your_api_token_here
Getting ClickUp Credentials
API Token
- Go to ClickUp Settings (click your avatar → Settings)
- Navigate to "Apps" in the sidebar
- Click "Generate" under "API Token"
- Copy the generated token
Usage
Development Mode
npm run dev
Production Build
npm run build
npm start
MCP Integration
Claude Desktop Configuration
Add this server to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"clickup": {
"command": "node",
"args": ["/path/to/clickup-mcp/dist/index.js"],
"env": {
"CLICKUP_API_TOKEN": "your_api_token_here"
}
}
}
}
Or for development:
{
"mcpServers": {
"clickup": {
"command": "npx",
"args": ["tsx", "/path/to/clickup-mcp/src/index.ts"],
"env": {
"CLICKUP_API_TOKEN": "your_api_token_here"
}
}
}
}
Typical Workflow
The server supports a hierarchical discovery flow:
Workspaces → Spaces → Folders → Lists → Tasks
Example Conversation
User: "What tasks do I have due today?"
AI:
-
First, let me discover your ClickUp structure...
→ list_workspaces Found: "Acme Corp" (workspace_id: "123") -
Searching for today's tasks across the workspace...
→ list_today_tasks({ workspace_id: "123" }) Found 5 tasks due today
User: "Create a new task in my Engineering sprint backlog"
AI:
-
Let me find that list...
→ list_spaces({ workspace_id: "123" }) Found: "Engineering" (space_id: "456") → list_lists({ space_id: "456" }) Found: "Sprint Backlog" (list_id: "789") -
Creating the task...
→ create_task({ list_id: "789", name: "New Task" }) Created successfully!
Tool Reference
Discovery Tools
list_workspaces
Lists all workspaces/teams you have access to.
Input: None
Output:
{
"count": 2,
"workspaces": [
{
"id": "123456",
"name": "My Workspace",
"color": "#7B68EE",
"members_count": 5
}
]
}
list_spaces
Lists all spaces in a workspace.
Input:
{
"workspace_id": "123456",
"include_archived": false
}
| Field | Type | Required | Description |
|---|---|---|---|
| workspace_id | string | Yes | Workspace ID |
| include_archived | boolean | No | Include archived spaces (default: false) |
list_folders
Lists all folders in a space.
Input:
{
"space_id": "789",
"include_archived": false
}
| Field | Type | Required | Description |
|---|---|---|---|
| space_id | string | Yes | Space ID |
| include_archived | boolean | No | Include archived folders (default: false) |
list_lists
Lists all lists in a space or folder.
Input:
{
"space_id": "789"
}
or
{
"folder_id": "456"
}
| Field | Type | Required | Description |
|---|---|---|---|
| space_id | string | One required | Space ID (gets all lists including folderless) |
| folder_id | string | One required | Folder ID (gets only lists in folder) |
| include_archived | boolean | No | Include archived lists (default: false) |
search_tasks
Search tasks across a workspace with optional filters.
Input:
{
"workspace_id": "123456",
"query": "bug fix",
"space_ids": ["789"],
"statuses": ["in progress"],
"include_closed": false
}
| Field | Type | Required | Description |
|---|---|---|---|
| workspace_id | string | Yes | Workspace ID |
| query | string | No | Search query |
| space_ids | string[] | No | Filter by spaces |
| folder_ids | string[] | No | Filter by folders |
| list_ids | string[] | No | Filter by lists |
| statuses | string[] | No | Filter by statuses |
| assignee_ids | number[] | No | Filter by assignees |
| include_closed | boolean | No | Include closed tasks |
Task Tools
list_today_tasks
Lists all tasks due today.
Input:
{
"workspace_id": "123456"
}
or
{
"list_id": "789"
}
| Field | Type | Required | Description |
|---|---|---|---|
| workspace_id | string | One required | Search across workspace |
| list_id | string | One required | Search specific list |
list_overdue_tasks
Lists all overdue tasks. Same input as list_today_tasks.
list_no_due_date_tasks
Lists all tasks without a due date in a specific list.
Input:
{
"list_id": "789"
}
| Field | Type | Required | Description |
|---|---|---|---|
| list_id | string | Yes | List ID |
create_task
Creates a new task in a specific list.
Input:
{
"list_id": "789",
"name": "Task Name",
"description": "Optional description",
"due_date": "2024-12-31T23:59:59Z",
"priority": "2",
"assignee": 12345678
}
| Field | Type | Required | Description |
|---|---|---|---|
| list_id | string | Yes | List ID to create task in |
| name | string | Yes | Task name |
| description | string | No | Task description |
| due_date | string | No | ISO 8601 date |
| priority | string | No | 1=Urgent, 2=High, 3=Normal, 4=Low |
| assignee | number | No | ClickUp user ID |
update_task
Updates an existing task.
Input:
{
"task_id": "abc123",
"status": "complete",
"due_date": "2024-12-31T23:59:59Z",
"priority": "1"
}
| Field | Type | Required | Description |
|---|---|---|---|
| task_id | string | Yes | Task ID to update |
| status | string | No | New status |
| due_date | string | No | New due date (ISO 8601) |
| priority | string | No | New priority |
add_comment
Adds a comment to a task.
Input:
{
"task_id": "abc123",
"comment": "This is a comment"
}
| Field | Type | Required | Description |
|---|---|---|---|
| task_id | string | Yes | Task ID |
| comment | string | Yes | Comment text |
Error Handling
The server provides detailed error responses:
{
"error": "ClickUp API error",
"status_code": 401,
"message": "ClickUp API error: 401 Unauthorized",
"details": { ... }
}
Common error scenarios:
- 401: Invalid API token
- 404: Task, list, folder, space, or workspace not found
- 400: Invalid request parameters
- Validation error: Invalid input format
Project Structure
clickup-mcp/
├─ src/
│ ├─ index.ts # MCP server entry point
│ ├─ clickup.ts # ClickUp API wrapper
│ ├─ tools.ts # MCP tool definitions
├─ dist/ # Compiled JavaScript (after build)
├─ package.json
├─ tsconfig.json
├─ .env.example
└─ README.md
Development
Type Checking
npx tsc --noEmit
Building
npm run build
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。