Airtable MCP Server
by domdomegg
Read and write Airtable bases directly from your AI agent
productivity Node.js Intermediate Self-hostable No API key
β 421 stars π
Updated: today
Description
Airtable MCP Server gives your AI agent full access to your Airtable bases β listing tables, reading records, creating entries, updating fields, and deleting rows. If you use Airtable as a lightweight database or project management tool, this server turns your AI into a hands-free Airtable operator that can query and modify data conversationally.
At version 1.9.6, this is one of the more mature MCP servers in the ecosystem. It supports all core Airtable operations and handles pagination, field types, and linked records correctly. The server also ships as a Docker image, which makes it easy to deploy in team environments. The codebase is clean with only 8 open issues, suggesting solid reliability.
The most practical use case is combining Airtable as a structured data layer with AI reasoning. Your agent can look up records, cross-reference data across tables, generate reports, and even maintain data quality by identifying and fixing inconsistencies β all through natural language. The limitation is that complex Airtable features like automations, interfaces, and scripting extensions are not exposed through MCP.
β Best for
Teams using Airtable as a data backend who want AI-powered data management
βοΈ Skip if
You need full Airtable automation features β use Airtable's built-in scripting instead
π‘ Use cases
- Query Airtable records using natural language instead of building views and filters manually
- Automate data entry by having your agent create and update records based on conversations
- Generate reports and summaries from Airtable data without exporting to spreadsheets
π Pros
- β Mature and stable (v1.9.6) with comprehensive CRUD operations
- β Available as npm package and Docker image for flexible deployment
- β Low issue count (8) indicates good reliability and maintenance
π Cons
- β Only covers data operations β Airtable automations, interfaces, and scripts are not accessible
- β Requires Airtable API token with appropriate base access permissions
- β Rate limited by Airtable's API (5 requests/second) which can slow bulk operations
π‘ Tips & tricks
Create a dedicated Airtable personal access token scoped to only the bases your agent
needs. Use read-only scopes initially and add write access only for bases where you want
AI modifications. For large bases, be mindful of Airtable's 5 req/sec rate limit β
batch operations work better than individual record updates.
Quick info
- Author
- domdomegg
- License
- MIT
- Runtime
- Node.js
- Transport
- stdio
- Category
- productivity
- Difficulty
- Intermediate
- Self-hostable
- β
- API key
- No API key needed
- Docker
- β
- Version
- 1.9.6
- Updated
- Feb 22, 2026
Client compatibility
- β Claude Code
- β Cursor
- β VS Code Copilot
- β Gemini CLI
- β Windsurf
- β Cline
- β JetBrains AI
- β Warp
Platforms
π macOS π§ Linux πͺ Windows