Overview

The Airtable MCP Server enables AI assistants to programmatically manage Airtable bases, tables, fields, and records through the Model Context Protocol. It provides a specialized implementation that builds complex tables in stages, minimizing failures common with multi-step operations.

Created by:

Developed by felores

Key Features

🗂️

Complete Base Management

Create, list, and manage Airtable bases, tables, and field configurations

📝

Intelligent Record Operations

Create, read, update, and delete records with filtering and search capabilities

🔄

Staged Table Construction

Build complex tables in stages to reduce complexity-related failures

🎨

Rich Field Type Support

Support for 14+ field types including text, email, numbers, dates, and selects

Available Tools

Quick Reference

Tool	Purpose	Category
`list_bases`	List all accessible Airtable bases	Base Management
`list_tables`	List tables within a base	Base Management
`create_table`	Create new table with fields	Base Management
`update_table`	Modify table name or description	Base Management
`create_field`	Add new field to table	Field Management
`update_field`	Modify existing field	Field Management
`list_records`	Retrieve records from table	Record Operations
`create_record`	Insert new record	Record Operations
`update_record`	Modify existing record	Record Operations
`delete_record`	Remove record	Record Operations
`search_records`	Find records matching criteria	Record Operations
`get_record`	Get single record by ID	Record Operations

Detailed Usage

list_bases▶

Retrieve all Airtable bases accessible to the authenticated user.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "list_bases",
  arguments: {}
});

Returns base IDs, names, and permission levels.

list_tables▶

List all tables within a specified Airtable base.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "list_tables",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX"
  }
});

create_table▶

Create a new table with specified fields in an Airtable base.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "create_table",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    name: "Projects",
    description: "Project tracking table",
    fields: [
      {
        name: "Project Name",
        type: "singleLineText"
      },
      {
        name: "Status",
        type: "singleSelect",
        options: {
          choices: [
            { name: "Active" },
            { name: "Completed" }
          ]
        }
      }
    ]
  }
});

The staged construction approach reduces failures when creating complex tables.

update_table▶

Update a table's name or description.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "update_table",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    table_id: "tblXXXXXXXXXXXXXX",
    name: "Updated Projects",
    description: "Updated description"
  }
});

create_field▶

Add a new field to an existing table.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "create_field",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    table_id: "tblXXXXXXXXXXXXXX",
    name: "Due Date",
    type: "date",
    options: {
      dateFormat: {
        name: "us"
      }
    }
  }
});

list_records▶

Retrieve records from a table with optional filtering, sorting, and pagination.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "list_records",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    table_id: "tblXXXXXXXXXXXXXX",
    view: "Grid view",
    maxRecords: 100
  }
});

create_record▶

Create a new record in a specified table.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "create_record",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    table_id: "tblXXXXXXXXXXXXXX",
    fields: {
      "Project Name": "New Website",
      "Status": "Active",
      "Due Date": "2025-12-31"
    }
  }
});

update_record▶

Modify an existing record's fields.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "update_record",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    table_id: "tblXXXXXXXXXXXXXX",
    record_id: "recXXXXXXXXXXXXXX",
    fields: {
      "Status": "Completed"
    }
  }
});

delete_record▶

Permanently delete a record from a table.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "delete_record",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    table_id: "tblXXXXXXXXXXXXXX",
    record_id: "recXXXXXXXXXXXXXX"
  }
});

This action cannot be undone.

search_records▶

Find records matching specific criteria using Airtable formulas.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "search_records",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    table_id: "tblXXXXXXXXXXXXXX",
    filterByFormula: "{Status} = 'Active'"
  }
});

get_record▶

Retrieve a single record by its unique ID.

use_mcp_tool({
  server_name: "airtable",
  tool_name: "get_record",
  arguments: {
    base_id: "appXXXXXXXXXXXXXX",
    table_id: "tblXXXXXXXXXXXXXX",
    record_id: "recXXXXXXXXXXXXXX"
  }
});

Supported Field Types

The Airtable MCP server supports all standard Airtable field types:

Text: singleLineText, multilineText
Contact: email, phoneNumber, url
Numeric: number, currency, percent, duration
Date & Time: date, dateTime
Selection: singleSelect, multipleSelects
References: multipleRecordLinks, multipleLookupValues
Advanced: checkbox, rating, barcode, button

Installation

Prerequisites

Node.js 18+ with npm installed
Airtable Personal Access Token with required scopes:
- data.records:read and data.records:write
- schema.bases:read and schema.bases:write

Get Your API Token

Visit Airtable Personal Access Tokens
Click "Create token"
Name your token (e.g., "MCP Server")
Add required scopes: data.records:read, data.records:write, schema.bases:read, schema.bases:write
Add access to specific bases
Click "Create token" and copy the token

Configuration

{
  "mcpServers": {
    "airtable": {
      "command": "npx",
      "args": [
        "-y",
        "@felores/airtable-mcp-server"
      ],
      "env": {
        "AIRTABLE_API_KEY": "your_personal_access_token_here"
      }
    }
  }
}

Keep Token Secure:

Your Airtable Personal Access Token is sensitive. Never share it publicly or commit it to version control.

Common Use Cases

Project Management

Automate project tracking and status updates:

"Create a new project record in Airtable with name 'Website Redesign',
status 'Active', and assign it to the Design team"

CRM Management

Manage contacts and relationships:

"Add a new contact to the CRM base: John Smith, email [email protected],
company Acme Corp, and tag as 'Prospect'"

Content Calendar

Organize content creation workflows:

"List all blog posts scheduled for next month from the Content Calendar table,
filtered by status 'In Progress'"

Sources

Figma MCP Server

Figma MCP servers enable AI models to access Figma design files, extract components, retrieve design tokens, and automate design-to-code workflows for seamless developer handoff.

Sequential Thinking MCP Server: AI Step-by-Step Problem Solving

Sequential Thinking MCP servers enable AI models to perform structured, step-by-step problem-solving with support for thought revision, branching reasoning, and dynamic context management.

Perplexity AI MCP Server: Real-time Search & Reasoning for AI

Explore Perplexity AI MCP Server for real-time web search, advanced reasoning, and comprehensive research. Empower your AI with up-to-date information and detailed answers.

Airtable MCP Server