Airtable

Airtable's MCP server lets AI tools connect directly to your Airtable bases. Once connected, you can ask questions about your data, create and update records, and analyse information through conversation with AI assistants such as Claude, ChatGPT and Cursor.

Model Context Protocol (MCP) is an open standard that lets AI tools connect securely to external data sources. Through it, an AI assistant can query your Airtable data, carry out actions on your behalf, and present structured information. The server respects your existing Airtable permissions and security settings.

The page lists several things you can ask for. You can search and analyse data, create and update records, create bases, read interface pages, build forms, and work with automations. Some actions have limits. Creating records is capped at 10 records per request, calls are subject to standard Airtable API rate limits, and automations containing Script actions must be edited in the Airtable UI.

The server is for people who already keep their data in Airtable and use an MCP-compatible AI tool. Owners, creators and editors can set up the integration and manage access, while commenters and read-only users can read data from the bases they can access.

Setup is covered in Airtable's developer docs, which include quickstart guides for Claude, Claude Code and ChatGPT. You need an MCP-enabled AI tool, data in Airtable bases, and permission to use third-party integrations in your Airtable organisation. If your organisation restricts integrations, an admin may need to allowlist the MCP integration first. You choose which bases your assistant can reach, and you can add or remove them later from your third-party integrations page.

According to the FAQ, the MCP integration is included with your current Airtable plan. There are no extra Airtable charges for MCP usage, but existing API rate limits still apply, and records created through MCP count against your base limits.