A set of DevOps tools for managing Airtable bases using the command line. This tool helps you maintain multiple environments (ex. development and production) for your Airtable bases and track structural changes (like deleting a field in a table) using version control.
- Maintain separate development and production environments
- Safely test changes in development before deploying to production
- Know what to edit in production to reflect developement changes
- Maintain a history of your base's evolution
- Review and understand schema modifications through clear diffs
- Collaborate with team members on schema changes
- 🔎 Generate schema files representing the structure of Airtable bases.
- ✅ Validate schema files to ensure they follow the correct structure used by this CLI.
- 📋 Compare two schema files to see structural changes in a human-readable format.
- Duplicate your existing base
- Decide which one will be the Production and Development base
- Create a folder on your computer for each one
- Install the CLI
- Move it to a folder with an env file to setup your Airtable PAT
- Get the schema for both bases:
# Get development base schema ./airtable-devops get-schema -b <development_base_id> -f ./myBase/dev/schema.json # Get production base schema ./airtable-devops get-schema -b <production_base_id> -f ./myBase/prod/schema.json
- You should have this directory setup :
├── airtable-devops(.exe)
├── .env
└── myBase
├── dev
│ └── schema.json
└── prod
└── schema.json
- Make changes on your Development base
-
Before deploying changes to production, generate a diff of your changes:
./airtable-devops diff --old ./myBase/prod/schema.json --new ./myBase/dev/schema.json > ./myBase/deployments/vX.X.txt -
Make the changes listed in
deployments/vX.X.txton your Production base -
After deploying to production, update your Production base schema:
./airtable-devops get-schema -b <production_base_id> -f ./myBase/prod/schema.json
- Visit the releases page
- Download the appropriate binary for your operating system
- Make the binary executable (Unix/Linux/macOS):
chmod +x airtable-devops
- Move the binary to a directory in your PATH (optional)
If you prefer to run the tool directly with Deno:
- Install Deno
- Clone this repository:
git clone https://github.com/Squix/airtable-devops.git cd airtable-devops
The get-schema command requires to interact with the Airtable API, hence you need to supply a valid Airtable Personal Access Token (PAT) that grants access to the resources you want the CLI to process.
You can provide your PAT in one of two ways:
-
Using a
.envfile in your project root with this content (recommended):AIRTABLE_PAT=your_airtable_pat
-
Exporting the variable in your shell:
# Bash export AIRTABLE_PAT=your_airtable_pat # Powershell $env:AIRTABLE_PAT=your_airtable_pat
The tool will automatically load the PAT from your .env file if present. If not found, it will fall back to checking the shell environment variable.
Generate schema files representing the structure of an Airtable base.
./airtable-devops get-schema --base-id <your_base_id> [--file <output_file> | --output-dir <output_directory>]--base-id(required): The base ID to get the schema from.--file(recommended): The file path where to save the schema. Ideal for version control as it can overwrites the same file.--output-dir(optional): The directory to save the schema file. Will create a new file each time. Default is./output/.
When using --output-dir, created schema files names follow this format : {baseId}_schema_{currentISOdate}.json
Example:
# Using --file (recommended for version control)
./airtable-devops get-schema --base-id app1234567890 --file ./schemas/my_base_schema.json
# Using --output-dir (creates timestamped files)
./airtable-devops get-schema --base-id app1234567890 --output-dir ./schemas/Validate a schema file to ensure it follows the correct structure for an Airtable base.
./airtable-devops validate --file <path_to_schema_file>--file(required): The path to the schema file to validate.
Example:
./airtable-devops validate --file ./schemas/base_schema.jsonThe command will provide detailed error messages if the schema is invalid, showing exactly what needs to be fixed and where.
Compare two schema files and show structural changes in a human-readable format.
./airtable-devops diff --old <old_schema_file> --new <new_schema_file> [--format <format>] [--color]--old(required): Path to the old schema file.--new(required): Path to the new schema file.--format(optional): Output format (text, json). Default is "text".--color(optional): Use colors in the output. Default is true.
Example:
./airtable-devops diff --old ./schemas/base_schema_v1.json --new ./schemas/base_schema_v2.jsonThe command will show a human-readable diff of the changes between the two schema files, including:
- Created, modified, and deleted tables
- Created, modified, and deleted fields
- Changes to field properties (name, type, description, options)
Example output:
Base: app1234567890
+ Projects (NEW TABLE)
+ Name (single_line_text)
+ Status (single_select)
~ Customers (MODIFIED TABLE)
~ Phone
type: text → phone_number
+ Email (email)
~ Address
type: single_line_text → multiline_text
options: + maxLength: 1000
- ZipCode (REMOVED)
- Archives (REMOVED TABLE)
If you're running the tool directly with Deno, you'll need to include the appropriate permission flags:
# Get Schema command
deno run --allow-net --allow-env --allow-write --allow-read main.ts get-schema --base-id <your_base_id> [--file <output_file> | --output-dir <output_directory>]
# Validate Schema command
deno run --allow-read main.ts validate --file <output_file>
# Diff Schema command
deno run --allow-read main.ts diff --old <old_schema_file> --new <new_schema_file> [--format <format>] [--color]Breakdown of needed Deno permissions:
| Permission flag | Usage |
|---|---|
| --allow-net | Access the Airtable REST API. |
| --allow-env | Load Airtable PAT from environment. |
| --allow-write | Write schema to a json file on disk. |
| --allow-read | Parse schema from a json file on disk. And tries to read PAT from .env file. |
Contributions are welcome! Please open an issue first for any improvements or bug fixes.
This project is licensed under the MIT License. See the LICENSE file for details.