Google Tasks MCP Server
Enables AI assistants to manage Google Tasks through natural language interactions. Supports creating, updating, deleting, searching, and listing tasks with secure OAuth2 authentication.
README
Google Tasks MCP Server
A Model Context Protocol (MCP) server for Google Tasks integration in Claude Desktop. This server enables AI assistants to manage Google Tasks through natural language interactions.
Forked from Zach Caceres' (zcaceres) original implementation to enhance authentication handling, add support for multiple task lists, and provide a more robust npm package with Smithery integration for Claude Desktop. This fork is fully functional and properly deployed on Smithery, making it ready for immediate use with Claude.
Features
- List all tasks in your Google Tasks lists
- Search for specific tasks by query
- Create new tasks with title, notes, and due date
- Update existing tasks (title, notes, status, due date)
- Delete tasks
- Clear completed tasks from a task list
- Full integration with Google Tasks API
- Secure OAuth2 authentication
- Docker support for containerized deployment
Task Creation Example
Here's a simple example of task creation using natural language with Claude:

With just a simple instruction like "create a task named 'eat a banana in the morning'", Claude can instantly create the task in your Google Tasks.
Installation
Installing via Smithery
To install Google Tasks Integration for Claude Desktop automatically:
npx -y @smithery/cli install @alvincrave/gtasks-mcp --client claude
Manual Installation
npm install @alvincrave/gtasks-mcp
Setup
-
Create a Google Cloud Project and obtain credentials:
a. Create a Google Cloud Project:
- Go to Google Cloud Console
- Create a new project or select an existing one
- Enable the Google Tasks API for your project
b. Create OAuth 2.0 Credentials:
- Go to "APIs & Services" > "Credentials"
- Click "Create Credentials" > "OAuth client ID"
- Choose "Desktop app" as application type
- Give it a name and click "Create"
- You will get your
GOOGLE_CLIENT_IDandGOOGLE_CLIENT_SECRET
c. Get Refresh Token:
- Go to OAuth 2.0 Playground
- Click the gear icon (Settings) in the top right
- Check "Use your own OAuth credentials"
- Enter your OAuth Client ID and Client Secret
- In the left panel, find "Tasks API v1" and select "https://www.googleapis.com/auth/tasks"
- Click "Authorize APIs" and complete the OAuth flow
- Click "Exchange authorization code for tokens"
- Copy the "Refresh token" - this is your
GOOGLE_REFRESH_TOKEN
-
Configure in Claude Desktop:
{
"mcpServers": {
"gtasks": {
"command": "npx",
"args": ["-y", "@alvincrave/gtasks-mcp"],
"env": {
"GOOGLE_CLIENT_ID": "your_client_id_here",
"GOOGLE_CLIENT_SECRET": "your_client_secret_here",
"GOOGLE_REFRESH_TOKEN": "your_refresh_token_here"
},
"connectionTypes": ["stdio"],
"autoRestart": true
}
}
}
Docker Support
You can run this server in a Docker container:
# Build the Docker image
docker build -t gtasks-mcp .
# Run the container
docker run -p 3000:3000 \
-e GOOGLE_CLIENT_ID=your_client_id \
-e GOOGLE_CLIENT_SECRET=your_client_secret \
-e GOOGLE_REFRESH_TOKEN=your_refresh_token \
gtasks-mcp
Usage Examples
The server provides several tools that can be used through Claude Desktop:
Search Tasks
search for tasks containing "meeting"
List Tasks
show me all my tasks
Create Task
create a task to "Prepare presentation" due on Friday
Update Task
mark the "Send email" task as completed
Delete Task
delete the task about "old project"
Clear Completed Tasks
clear all completed tasks from my list
Security Notes
- Keep your Google API credentials secure
- Regularly rotate your refresh tokens
- Store sensitive information in Claude Desktop configuration
- Never share or commit your credentials to version control
- The refresh token gives access to your Google Tasks, treat it like a password
License
This project is licensed under the MIT License.
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。