Lists all tables in the connected Airtable base
Retrieves the schema (field structure) of a specific table
Lists records from a specified table with optional filtering
Retrieves a specific record by its ID
Creates a new record in a specified table
Updates an existing record in a specified table
Deletes a record from a specified table
The Airtable Integration enables AI-powered applications to interact with Airtable databases through Anthropic's Model Context Protocol (MCP). This integration allows AI tools to query, create, update, and delete Airtable records using natural language commands. With features including base management, table operations, schema manipulation, record filtering, and data migration, this tool bridges the gap between AI systems and structured data storage. Compatible with Claude Desktop and other Claude-powered editors, it provides a standardized interface for AI to leverage the power of Airtable's database capabilities.
The Airtable Integration MCP allows AI tools to interact with Airtable databases through a standardized interface. This integration enables seamless data access and management directly from AI applications, making it possible to query, manipulate, and analyze Airtable data using natural language.
Clone the repository
git clone https://github.com/rashidazarang/airtable-mcp.git
cd airtable-mcp
Install dependencies
For JavaScript implementation:
npm install
For Python implementation:
pip install -r requirements.txt
Get your Airtable credentials
appi7fWMQcB3BNzPs
)To configure Claude Desktop to use the Airtable MCP:
Open the Claude Desktop configuration file:
~/Library/Application Support/Claude/claude_desktop_config.json
Add the Airtable MCP configuration:
{
"mcpServers": {
"airtable": {
"command": "node",
"args": [
"/path/to/airtable-mcp/airtable_simple.js",
"--api-key",
"YOUR_AIRTABLE_API_KEY",
"--base-id",
"YOUR_BASE_ID"
]
}
}
}
Replace /path/to/airtable-mcp
with the actual path to the cloned repository, and add your Airtable API key and base ID.
You can also configure the MCP using a URL parameter:
{
"mcpServers": {
"airtable": {
"url": "http://localhost:3333?apiKey=YOUR_AIRTABLE_API_KEY&baseId=YOUR_BASE_ID"
}
}
}
For the Python implementation, you can run the server directly:
python simple_airtable_server.py --api-key YOUR_AIRTABLE_API_KEY --base-id YOUR_BASE_ID
Once configured, you can interact with your Airtable bases through Claude. Here are some example queries:
The MCP handles translating these natural language requests into appropriate Airtable API calls and returns the results in a structured format.
The repository includes a Dockerfile for containerized deployment:
docker build -t airtable-mcp .
docker run -p 3333:3333 -e AIRTABLE_API_KEY=your_api_key -e AIRTABLE_BASE_ID=your_base_id airtable-mcp
For Smithery users, the repository includes a smithery.yaml
file that can be used to configure the MCP:
smithery deploy @rashidazarang/airtable-mcp
If you encounter issues:
For more detailed troubleshooting, refer to the repository's issue templates or create a new issue on GitHub.