Airtable Integration MCP Server

Overview

The Airtable Integration MCP server allows AI assistants to interact with your Airtable databases. This enables your AI assistant to perform database operations through natural language, making it easier to manage and manipulate your data without switching contexts.

Installation

To use the Airtable Integration with your AI assistant, you'll need to add it to your configuration file. The server requires an Airtable personal access token to authenticate with your account.

For Claude Desktop

Add the following to the "mcpServers" section of your claude_desktop_config.json:

{
  "mcpServers": {
    "airtable": {
      "command": "npx",
      "args": [
        "-y",
        "airtable-mcp-server"
      ],
      "env": {
        "AIRTABLE_API_KEY": "YOUR_AIRTABLE_PAT_HERE"
      }
    }
  }
}

Replace YOUR_AIRTABLE_PAT_HERE with your Airtable personal access token. Your token should have at least schema.bases:read and data.records:read permissions, and optionally the corresponding write permissions if you want to modify data.

Using Docker

If you prefer to use Docker, you can run the server with:

docker run -e AIRTABLE_API_KEY=YOUR_AIRTABLE_PAT_HERE ghcr.io/domdomegg/airtable-mcp-server

Usage

Once installed, your AI assistant can interact with your Airtable bases. Here are some example interactions:

1. Exploring your databases:

   • "Show me all my Airtable bases"
   • "List the tables in my Project Management base"
   • "Describe the structure of my Tasks table"

2. Querying data:

   • "Find all records in the Customers table that contain 'Smith'"
   • "Get the details of record rec123 from the Orders table"
   • "List the first 10 records from the Inventory table"

3. Modifying data:

   • "Create a new record in the Tasks table with title 'Review proposal' and status 'Pending'"
   • "Update the status of task rec456 to 'Completed'"
   • "Delete the expired records from the Events table"

4. Modifying structure (if you have appropriate permissions):

   • "Create a new table called 'Projects' with fields for name, description, and deadline"
   • "Add a new field called 'Priority' to the Tasks table"

The AI will use the appropriate tools to fulfill your requests, translating your natural language instructions into the corresponding Airtable API operations.

Permissions

The operations available to the AI assistant depend on the permissions granted to your Airtable personal access token. For read-only access, use the following scopes:

• schema.bases:read
• data.records:read

For full read/write access, add these scopes:

• schema.bases:write
• data.records:write

You can create and manage your tokens at https://airtable.com/create/tokens.