Nylas API MCP Server
Enables to interact with Nylas API for email, calendar, and contacts via MCP, providing documentation, code samples, and code generation tools.
README
Nylas API MCP Server (Experimental)
Note: This project is experimental and is intended as an exploration of using the Model Context Protocol (MCP) as a guide for Nylas API integrations. It is not official and should be used for learning and experimentation purposes only.
This project implements a Model Context Protocol (MCP) server for the Nylas API. It provides resources, tools, and prompts to help developers learn about and integrate with the Nylas API for email, calendar, and contacts functionality.
What is MCP?
The Model Context Protocol (MCP) is a standard for integrating data, tools, and prompts with AI applications. MCP servers can expose resources, tools, and prompts to AI applications like Claude Desktop, Cursor, and other MCP-compatible applications.
Features
This MCP server provides:
- Documentation Resources: Comprehensive documentation about Nylas API endpoints, authentication, and best practices
- Code Samples: Ready-to-use code samples for various Nylas API operations in multiple languages (Node.js, Python, Java, Ruby, curl)
- Interactive Tools: Tools to generate authentication code and endpoint-specific implementation code
- Prompts: Pre-built prompts for common Nylas integration scenarios
Getting Started
Prerequisites
- Node.js 18+
- npm
Installation
- Clone this repository
- Install dependencies
npm install
- Build the project
npm run build
Running the MCP Server
Using stdio (for direct integration with MCP clients)
npm start
Using HTTP with SSE (for remote connections)
MCP_MODE=http npm start
By default, the HTTP server runs on port 3000. You can change this by setting the PORT environment variable:
MCP_MODE=http PORT=8080 npm start
Using with MCP Clients
Claude Desktop
- Start the MCP server in stdio mode
- In Claude Desktop, go to Settings > Servers
- Click "Add Server"
- Select "Run a local command" and enter the command to start this server:
node /path/to/nylas-api-mcp/dist/index.js
- Give it a name (e.g., "Nylas API")
- After adding, you can use all the Nylas API resources, tools, and prompts in your Claude conversations
Cursor
- Start the MCP server in stdio mode
- In Cursor, add a new MCP server in the settings
- Configure it to use the command:
node /path/to/nylas-api-mcp/dist/index.js
Resources
This MCP server provides the following resources:
- General Documentation: Overview of Nylas API capabilities
- Authentication Guide: How to authenticate with Nylas API
- API Endpoints: Documentation for email, calendar, contacts, and webhook endpoints
- Code Samples: Usage examples in multiple programming languages
Tools
Interactive tools for code generation:
- generate-auth-code: Generates authentication code in your preferred language
- generate-endpoint-code: Generates code for specific API endpoints
- search-api-docs: Search through the Nylas API documentation
Prompts
Pre-built prompts for common scenarios:
- Getting Started: Basic intro to Nylas API
- Authentication Guide: Understanding OAuth flow
- Email/Calendar/Contacts Integration: Feature-specific guidance
- Integration Scenarios: Guidance for specific use cases (email client, calendar booking, etc.)
- SDK Examples: Example code in various languages
- Debugging Guide: Common issues and solutions
Contributing
Contributions are welcome! You can:
- Add more code samples to the
nylas-code-samplesdirectory - Improve documentation resources
- Add support for additional programming languages
- Create new tools and prompts
Disclaimer
This is an experimental project and is not officially supported by Nylas. The information and code provided through this MCP server should be used as a learning resource only. Always refer to the official Nylas documentation for the most accurate and up-to-date information on the Nylas API.
The sample code provided is for educational purposes and may need additional error handling, security considerations, and testing before being used in production environments.
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。