ticktick-mcp-server

ticktick-mcp-server

An MCP server that enables Claude to manage TickTick tasks and projects via OAuth 2.1 with PKCE authentication.

Category
访问服务器

README

ticktick-mcp-server

日本語版はこちら

TickTick MCP (Model Context Protocol) Server - An MCP server for using TickTick task management features from Claude

Overview

This MCP server enables integration with the TickTick task management platform from Claude Desktop. It implements secure authentication using OAuth 2.1 with PKCE (Proof Key for Code Exchange) and provides comprehensive task and project management capabilities.

Authentication and Authorization

OAuth 2.1 with PKCE

This server adopts the OAuth 2.1 with PKCE protocol for secure authentication:

  1. PKCE Authentication Flow

    • Generates code_verifier and code_challenge at the start of authentication
    • Displays TickTick authentication screen in browser
    • Obtains authorization code after user approval
    • Exchanges tokens using authorization code and code_verifier
  2. Token Storage

    • Stores access tokens in .ticktick-mcp-server-credentials.json file
    • ⚠️ Security Warning: Tokens are currently stored in plaintext. For production use, implementing encryption is recommended
    • Never expose access tokens publicly
  3. Automatic Token Refresh

    • MCP server automatically checks token expiration
    • Obtains new access token using refresh token when expired

Setup Instructions

1. Configure Authentication Credentials

Configure credentials obtained from the TickTick Developer Console. Choose from one of three methods:

Method 1: ticktick-oauth.keys.json (Recommended)

Create a ticktick-oauth.keys.json file in the project root:

{
  "client_id": "your_ticktick_client_id",
  "client_secret": "your_ticktick_client_secret",
  "redirect_uri": "http://localhost:8080/callback"
}

Method 2: Environment Variables (.env)

Create a .env file in the project root:

TICKTICK_CLIENT_ID=your_ticktick_client_id
TICKTICK_CLIENT_SECRET=your_ticktick_client_secret
TICKTICK_REDIRECT_URI=http://localhost:8080/callback

Method 3: System Environment Variables

Set directly in your shell:

export TICKTICK_CLIENT_ID=your_ticktick_client_id
export TICKTICK_CLIENT_SECRET=your_ticktick_client_secret
export TICKTICK_REDIRECT_URI=http://localhost:8080/callback

Configuration Priority:

  1. Settings passed directly to the constructor (specified in code)
  2. ticktick-oauth.keys.json file
  3. Environment variables (.env file or system environment variables)

Note:

  • When configured through multiple methods, loading follows the priority order above
  • For security, these files are added to .gitignore and will not be committed to the repository
  • Replace your_ticktick_client_id and your_ticktick_client_secret with actual values

Token Storage Location Configuration (Important)

Authentication tokens are saved according to the following priority:

  1. Path specified by TICKTICK_CREDENTIALS_PATH environment variable (recommended)
  2. User home directory (~/.ticktick-mcp-server-credentials.json)
  3. Current working directory (fallback)

For Claude Desktop environment, setting the TICKTICK_CREDENTIALS_PATH environment variable is strongly recommended:

export TICKTICK_CREDENTIALS_PATH=/Users/yourusername/Documents/.ticktick-mcp-server-credentials.json

2. Build the Project

npm install
npm run build

For Developers: Testing and Code Quality

# TypeScript type checking
npm run typecheck

# Code quality check with ESLint
npm run lint

# Run tests
npm test

# Check test coverage (target: 80% or higher)
npm run test:coverage

# Run tests in watch mode
npm run test:watch

3. Execute Authentication

For first-time use, authenticate with the following command:

# Start authentication flow
npm run start auth
# or
npm run start:auth

Authentication Steps

  1. Running the above command will automatically open a browser
  2. Log in to your account on the TickTick login screen
  3. Approve application access permissions
  4. Upon completion, tokens are saved to .ticktick-mcp-server-credentials.json
  5. "Authentication completed successfully!" message will be displayed

4. Verify Authentication Status

To verify authentication has completed successfully:

# Launch MCP Inspector to verify
npm run inspector
  1. Access http://localhost:5173 in your browser
  2. Click "Connect" in MCP Inspector
  3. Execute the ticktick_auth_status tool to check authentication status

5. Normal Startup

After authentication is complete, start with:

npm run start

Available MCP Tools

Authentication Management

  • ticktick_auth_status: Check authentication status
  • ticktick_authorize: Start OAuth authentication flow
  • ticktick_logout: Clear saved tokens
  • ticktick_health_check: Test API connectivity

Task Management

  • ticktick_list_tasks: Get task list (with filtering and sorting)
  • ticktick_get_task: Get specific task details
  • ticktick_create_task: Create new task
  • ticktick_update_task: Update task (title, content, due date, priority, checklist, etc.)
    • Note: projectId and taskId parameters are required (per TickTick API specification)
  • ticktick_delete_task: Delete task
  • ticktick_complete_task: Mark task as completed
  • ticktick_uncomplete_task: Mark task as incomplete
  • ticktick_search_tasks: Search tasks
  • ticktick_get_overdue_tasks: Get overdue tasks

Project Management

  • ticktick_list_projects: Get project list
  • ticktick_get_project: Get specific project details
  • ticktick_create_project: Create new project
  • ticktick_update_project: Update project
  • ticktick_delete_project: Delete project
  • ticktick_get_project_tasks: Get tasks in a project
  • ticktick_get_project_stats: Get project statistics
  • ticktick_archive_project: Archive project
  • ticktick_unarchive_project: Unarchive project

API Features

This MCP server is implemented based on the official TickTick OpenAPI:

Supported Operations

  • Create, update, delete, and complete tasks
  • Create, update, and delete projects
  • Project archive functionality
  • Task search and filtering

Implementation Features

  • Automatic date format conversion: Auto-conversion from ISO format to TickTick API format
  • Enhanced error handling: Detailed error messages and troubleshooting hints
  • Rate limiting support: API client with automatic retry functionality
  • PKCE-enabled OAuth: Secure authentication flow

Integration with Claude Desktop

Add the following to your Claude Desktop configuration file:

{
  "mcpServers": {
    "ticktick": {
      "command": "/absolute/path/to/node",
      "args": ["/absolute/path/to/ticktick-mcp-server/dist/index.js"],
      "env": {
        "TICKTICK_CLIENT_ID": "your_client_id",
        "TICKTICK_CLIENT_SECRET": "your_client_secret",
        "TICKTICK_REDIRECT_URI": "http://localhost:8080/callback",
        "TICKTICK_CREDENTIALS_PATH": "/absolute/path/to/.ticktick-mcp-server-credentials.json",
        "LOG_LEVEL": "info"
      }
    }
  }
}

Troubleshooting

Authentication Errors

"invalid_request" Error

OAuth Error
error="invalid_request", error_description="At least one redirect_uri must be registered with the client."

Cause: No redirect URI is registered in the TickTick OAuth application.

Solution:

  1. Access TickTick Developer Portal
  2. Add the following redirect URI in application settings:
    http://localhost:8080/callback
    
  3. Save settings

Important: The redirect URI must match exactly (protocol, host, port, and path).

Other Authentication Errors

  1. Verify the contents of ticktick-oauth.keys.json are correct
  2. Confirm redirect_uri is set to http://localhost:8080/callback
  3. Check that the application is enabled in the TickTick Developer Console

Token Not Being Saved

  1. Check write permissions for the project directory
  2. Verify .ticktick-mcp-server-credentials.json file is created
  3. Confirm file contents are valid JSON format

Re-authenticating

  1. Delete .ticktick-mcp-server-credentials.json file
  2. Run npm run start auth command again

Latest Updates

See CHANGELOG.md for detailed change history.

v0.2.0 (Latest) - 2025-06-08

  • Comprehensive test suite added: Jest configuration and unit tests (80% coverage target)
  • Input validation implemented: Strict type checking and validation with Zod schemas
  • Unified error handling: Consistent error response format
  • ESLint configuration added: Strict code quality rules for TypeScript
  • Expanded documentation: API specification, CHANGELOG, and contribution guidelines added
  • Improved type safety: Eliminated any types with proper type annotations

Security

Important Security Notes

  1. Protecting Credentials

    • .ticktick-mcp-server-credentials.json - Contains authentication tokens
    • ticktick-oauth.keys.json - Contains OAuth client secrets
    • Never commit these files to Git
  2. Recommended Security Measures

    • Set appropriate permissions for credential files: chmod 600 .ticktick-mcp-server-credentials.json
    • Rotate tokens regularly
    • Delete unused tokens
    • Implement token encryption for production environments

See Security Policy for details.

Documentation

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选