Gmail MCP Server
A Model Context Protocol server that enables AI assistants to search, read, and send emails with attachment support through the Gmail API. It also provides tools for comprehensive label management and uses secure OAuth2 authentication for mailbox access.
README
Gmail MCP Server
A Model Context Protocol (MCP) server implementation for Gmail integration, enabling AI assistants to interact with Gmail through a standardized interface.
Features
-
Email Operations
- List and search emails with advanced filtering
- Read email content with attachments
- Send emails with attachments
- Draft support (coming soon)
- Reply/Forward support (coming soon)
-
Label Management
- List all labels
- Create new labels
- Update existing labels
- Delete labels
-
Authentication
- Secure OAuth2 authentication
- Automatic token refresh
- Token persistence
Installation
- Clone the repository:
git clone https://github.com/yourusername/gmail-mcp.git
cd gmail-mcp
- Install dependencies:
npm install
-
Set up Google Cloud Project:
- Go to Google Cloud Console
- Create a new project
- Enable Gmail API
- Configure OAuth consent screen
- Create OAuth credentials
- Download credentials as
gcp-oauth.keys.json
-
Configure the project:
- Place
gcp-oauth.keys.jsonin the project root directory - Run the authentication server:
- Place
npm run auth
- Build the project:
npm run build
Usage
Starting the Server
npm start
The server runs on stdio, making it compatible with MCP clients.
Available Tools
1. List Emails
{
"name": "list-emails",
"arguments": {
"maxResults": 10,
"labelIds": ["INBOX"],
"query": "is:unread"
}
}
2. Read Email
{
"name": "read-email",
"arguments": {
"id": "message-id",
"format": "full"
}
}
3. Send Email
{
"name": "send-email",
"arguments": {
"to": "recipient@example.com",
"subject": "Hello",
"body": "Message content",
"attachments": ["/path/to/file.pdf"]
}
}
4. Search Emails
{
"name": "search-emails",
"arguments": {
"query": "from:sender@example.com",
"maxResults": 5
}
}
5. Manage Labels
{
"name": "manage-labels",
"arguments": {
"action": "create",
"name": "MyNewLabel"
}
}
Configuration
Environment Variables
GMAIL_MCP_DEBUG: Enable debug logging (default: false)GMAIL_MCP_TOKEN_PATH: Custom path for token storageGMAIL_MCP_KEYS_PATH: Custom path for OAuth keys file
OAuth Credentials
The OAuth credentials file (gcp-oauth.keys.json) should be structured as:
{
"installed": {
"client_id": "your-client-id",
"client_secret": "your-client-secret",
"redirect_uris": ["http://localhost:3000/oauth2callback"]
}
}
Development
Running Tests
# Run all tests
npm test
# Run with coverage
npm run test:coverage
# Run in watch mode
npm run test:watch
Building
# Build once
npm run build
# Build in watch mode
npm run dev
Error Handling
The server handles various error scenarios:
- Authentication failures
- Rate limiting
- Invalid parameters
- Network issues
- Permission errors
Errors are returned in a standardized format with appropriate HTTP status codes.
Contributing
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
License
This project is licensed under the MIT License - see the LICENSE file for details.
Acknowledgments
- Google Gmail API
- Model Context Protocol
- Contributors and maintainers
Support
For support, please open an issue in the GitHub repository or contact the maintainers.
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。
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。
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。
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。
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。
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。
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。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。