Connect your AI tools directly to Airtable. Query, create, update, and delete records using natural language. Features include base management, table operations, schema manipulation, record filtering, and data migration—all through a standardized MCP interface compatible with Cursor, Claude Code, Cline, Zed, and other Claude-powered editors.
This application is a powerful Airtable integration tool that enables AI-powered applications via Anthropic's Model Context Protocol (MCP) to access and manipulate Airtable data directly from your IDE.
- Base Management: List and select Airtable bases
- Table Operations: Browse tables, fields, and records
- Data Access: Read, create, update, and delete records
- Schema Management: Export, compare, and update schemas
- Command-line Configuration: Use API tokens directly through command-line parameters
- NPX Compatible: Easy installation with a single command
- Smithery Integration: One-click installation via Smithery
There are two core components used to access and manipulate Airtable data:
- Airtable MCP Server: A Python server that provides standardized tools for AI clients to interact with Airtable.
- MCP Client: Any client that supports the Model Context Protocol (Cursor, Claude Desktop, Cline, Zed, etc.).
┌─────────────┐ ┌──────────────┐ ┌───────────────┐
│ MCP Client │ ──► │ Airtable │ ──► │ Airtable │
│ (e.g. │ ◄── │ MCP Server │ ◄── │ API │
│ Cursor) │ │ │ │ │
└─────────────┘ └──────────────┘ └───────────────┘
Model Context Protocol (MCP) is a capability supported by Anthropic AI models that allows you to create custom tools for any compatible client. MCP clients like Claude Desktop, Cursor, Cline, or Zed can run an MCP server which "teaches" these clients about new tools they can use.
- Node.js 14+
- Python 3.10+ (automatically detected)
- Airtable API token
- A compatible MCP client (Cursor, Claude Desktop, etc.)
The easiest way to install:
- Visit Smithery
- Search for "@rashidazarang/airtable-mcp"
- Click "Install" and follow the prompts
Another fast way to get started:
# Install globally
npm install -g airtable-mcp
# Or run directly with npx
npx airtable-mcp --token "your_airtable_token" --base "your_base_id"For Cursor, update your ~/.cursor/mcp.json file:
{
"mcpServers": {
"airtable-tools": {
"command": "npx",
"args": [
"airtable-mcp",
"--token", "your_airtable_token",
"--base", "your_base_id"
]
}
}
}Restart your MCP client to load the new tools.
If you prefer to clone the repository and install manually:
-
Clone this repository:
git clone https://github.com/rashidazarang/airtable-mcp.git cd airtable-mcp -
Install dependencies:
pip install -r requirements.txt
-
Run the server:
python airtable_mcp/src/server.py --token "your_airtable_token" --base "your_base_id"
Once installed and configured, you can use natural language to interact with your Airtable data:
- "Show me all the bases I have access to"
- "List the tables in my current base"
- "Show me the structure of the Customers table"
- "Get the first 10 records from the Orders table"
- "Create a new record in the Products table"
- "Update record XYZ in the Inventory table"
- "Export the schema of my current base"
| Tool Name | Description |
|---|---|
list_bases |
List all accessible Airtable bases |
list_tables |
List all tables in the specified or default base |
list_records |
List records from a table with optional filtering |
get_record |
Get a specific record from a table |
create_records |
Create records in a table from JSON string |
update_records |
Update records in a table from JSON string |
delete_records |
Delete records from a table by IDs |
export_records |
Export records from a table as JSON string |
import_records |
Import records to a table from JSON string |
create_tables |
Create tables from a JSON schema |
update_schema |
Update existing tables to match a JSON schema |
set_base_id |
Set the current Airtable base ID |
inspect_table |
Get detailed information about a table's structure and fields |
export_schema |
Export the schema of the current base in JSON or CSV format |
compare_schemas |
Compare the provided schema with the current base schema |
generate_field_mapping |
Generate a field mapping between two tables |
migrate_data |
Migrate data from one table to another using the specified field mapping |
For production use, you can use PM2 to manage the Airtable MCP server:
// ecosystem.config.js
module.exports = {
apps: [
{
name: 'airtable-mcp',
script: 'npx',
args: [
'airtable-mcp',
'--token', 'your_airtable_token',
'--base', 'your_base_id'
],
interpreter: '/usr/local/bin/node',
env: {
PATH: process.env.PATH,
},
},
],
};Start with:
pm2 start ecosystem.config.js- Works with any MCP-compatible client
- Primarily designed for Cursor IDE integration
- Supports other AI editors and MCP clients
For detailed installation instructions, troubleshooting, and advanced configurations, see INSTALLATION.md.
MIT
Contributions are welcome! Please feel free to submit a Pull Request.