Airtable MCP Server
Enables complete interaction with Airtable databases through 16 CRUD operations including batch processing, schema management, and record manipulation. Designed for AI applications and n8n workflows with HTTP streaming support.
README
Airtable MCP Server
A Model Context Protocol (MCP) server for Airtable, designed to integrate with AI-powered applications and n8n workflows. This server provides a standardized interface for interacting with Airtable bases, tables, and records.
🚀 Features
- 16 Core Tools: Complete CRUD operations for Airtable
- Batch Operations: Create, update, and delete multiple records at once
- Schema Management: Access table, field, and view information
- HTTP Streamable: Compatible with n8n and other HTTP-based MCP clients
- TypeScript: Full type safety and modern JavaScript features
- Render Ready: Pre-configured for deployment on Render
🛠️ Available Tools
Base Operations
get_base_info- Get information about the Airtable baselist_tables- List all tables in the base
Table Operations
get_table_info- Get detailed table informationlist_fields- List all fields in a tableget_field_info- Get detailed field information
Record Operations
list_records- List records with filtering, sorting, and paginationget_record- Get a specific record by IDcreate_record- Create a new recordupdate_record- Update an existing recorddelete_record- Delete a record
Batch Operations
create_records- Create multiple records (up to 10)update_records- Update multiple records (up to 10)delete_records- Delete multiple records (up to 10)
View Operations
list_views- List all views for a tableget_view_info- Get detailed view information
🚀 Quick Start
Prerequisites
- Node.js 20+
- Airtable API key
- Airtable base ID
Installation
-
Clone the repository
git clone https://github.com/jjwjr94/airtable-mcp.git cd airtable-mcp -
Install dependencies
npm install -
Set up environment variables
cp env.example .env # Edit .env with your Airtable credentials -
Start the development server
npm run dev:http
Environment Variables
# Server Configuration
NODE_ENV=development
PORT=3000
# Airtable Configuration
AIRTABLE_API_KEY=your_airtable_api_key_here
AIRTABLE_BASE_ID=your_base_id_here
🔧 Usage
HTTP Server (Recommended for n8n)
The HTTP server runs on port 3000 by default and provides these endpoints:
- Health Check:
GET /health - MCP Endpoint:
POST /mcp - Set Credentials:
POST /set-credentials
Headers Required
For all MCP requests, include these headers:
x-airtable-api-key: YOUR_API_KEY
x-airtable-base-id: YOUR_BASE_ID
Example MCP Request
curl -X POST http://localhost:3000/mcp \
-H "Content-Type: application/json" \
-H "x-airtable-api-key: YOUR_API_KEY" \
-H "x-airtable-base-id: YOUR_BASE_ID" \
-d '{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "list_tables",
"arguments": {}
},
"id": "1"
}'
🚀 Deployment
Render (Recommended)
- Fork this repository
- Connect to Render
- Deploy as a Web Service
- Set environment variables in Render dashboard
The render.yaml file is pre-configured for easy deployment.
Environment Variables for Production
NODE_ENV=production
PORT=10000
🧪 Testing
Run Tests
# Start the server first
npm run dev:http
# In another terminal, run tests
npm test
Test Individual Endpoints
# Health check
curl http://localhost:3000/health
# List tools
curl -X POST http://localhost:3000/mcp \
-H "Content-Type: application/json" \
-H "x-airtable-api-key: YOUR_API_KEY" \
-H "x-airtable-base-id: YOUR_BASE_ID" \
-d '{"jsonrpc":"2.0","method":"tools/list","params":{},"id":"1"}'
🔌 n8n Integration
HTTP Request Node Configuration
- Method: POST
- URL:
https://your-render-app.onrender.com/mcp - Headers:
Content-Type: application/jsonx-airtable-api-key: {{ $json.apiKey }}x-airtable-base-id: {{ $json.baseId }}
- Body:
{ "jsonrpc": "2.0", "method": "tools/call", "params": { "name": "list_tables", "arguments": {} }, "id": "{{ $json.requestId }}" }
Example n8n Workflow
- Set node to configure Airtable credentials
- HTTP Request node to call MCP tools
- Switch node to handle different tool responses
- Process node to format data for your needs
📚 API Reference
Tool: list_records
List records from a table with optional filtering and pagination.
Parameters:
tableId(required): The ID of the tablepageSize(optional): Number of records to return (max 100)filterByFormula(optional): Airtable formula to filter recordssort(optional): Sorting configurationfields(optional): Specific fields to returnview(optional): View ID to use
Example:
{
"name": "list_records",
"arguments": {
"tableId": "tblXXXXXXXXXXXXXX",
"pageSize": 50,
"filterByFormula": "{Status} = 'Active'",
"sort": [{"field": "Name", "direction": "asc"}],
"fields": ["Name", "Status", "Created"]
}
}
Tool: create_record
Create a new record in a table.
Parameters:
tableId(required): The ID of the tablefields(required): The field values for the new record
Example:
{
"name": "create_record",
"arguments": {
"tableId": "tblXXXXXXXXXXXXXX",
"fields": {
"Name": "New Project",
"Status": "Active",
"Priority": "High"
}
}
}
🔒 Security
- CORS: Configured for Render and n8n domains
- Helmet: Security headers enabled
- Input Validation: All inputs are validated
- Error Handling: Secure error messages
🚀 Development
Scripts
npm run build # Build TypeScript
npm run start # Start production server
npm run start:http # Start HTTP server
npm run dev # Development mode
npm run dev:http # HTTP development mode
npm run watch # Watch mode for TypeScript
npm test # Run tests
npm run inspector # MCP inspector
Project Structure
airtable-mcp/
├── src/
│ ├── airtable-client-wrapper.ts # Airtable API client
│ ├── tool-handler.ts # MCP tool definitions
│ └── version.ts # Version information
├── index.ts # Standard MCP server
├── index-http.ts # HTTP server
├── test-http-server.js # Test suite
├── render.yaml # Render deployment config
└── package.json # Dependencies and scripts
🤝 Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
📄 License
MIT License - see LICENSE file for details.
🙏 Acknowledgments
- Built for the Model Context Protocol
- Powered by Airtable API
- Compatible with Claude Desktop and other MCP clients
- Designed for n8n integration
📞 Support
- Issues: GitHub Issues
- Discussions: GitHub Discussions
Version: 1.0.0 | Status: 🚀 Production Ready | MCP Protocol: 2025-06-18 | Type Safety: Full TypeScript | Last Updated: January 2025
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。