The Scenario
You are a data engineer who just joined a company. Your first week task: document the main Airtable base, called Operations Hub, before you build any pipelines on top of it.
The base has 8 tables. You need a data dictionary showing every table, every field name, every field type, and any description attached to the field. You need this in an Excel workbook that the BI lead can annotate and use to spec the warehouse schema.
What documenting this manually looks like:
- You open Airtable, click the first table, open the field editor for each field
- You copy the field name, field type, and description into a workbook row by row
- Eight tables, forty-plus fields per table on average — Wednesday afternoon is gone
- You finish, send it over, and the BI lead replies: the Linked Records fields are missing their linked table reference
- You have to go back through 8 tables and check every linked field.
The fast version is one prompt.
The Easy Way: One Prompt in SheetXAI
SheetXAI calls the Airtable metadata API, reads the full schema of every table, and writes it into the workbook, including linked table references.
Open the SheetXAI sidebar and type:
Fetch the full schema of my Airtable base 'Operations Hub' and write each table name, field name, and field type as a row in this workbook. Use the Schema tab. Add a header row with columns: Table, Field Name, Field Type, Description. Include hidden fields and for linked record fields include the name of the linked table.
SheetXAI calls the Airtable base schema endpoint, reads all 8 tables and every field, and writes the data dictionary into the Schema tab.
What You Get
A complete data dictionary in the workbook:
- One row per field — table name, field name, field type, description
- All 8 tables covered — nothing skipped because of view filters
- Hidden fields included — the metadata API returns all fields regardless of view visibility
- Linked table references — for multipleRecordLinks fields, the linked table name is included
- Field types in Airtable's vocabulary — singleLineText, date, singleSelect, multipleRecordLinks, etc.
The whole thing takes as long as it takes the API to respond, not a Wednesday afternoon of clicking.
Once the dictionary is in the workbook, you can sort by field type to find all date fields, filter to a single table, or add an annotation column for the BI lead.
What If the Data Is Not Quite Ready
Schema documentation has its own edge cases. SheetXAI handles them in the same prompt.
When you have the Airtable URL but not the base ID
You have the base open in a browser tab and can see the URL, but the base ID is buried in it.
The Airtable base URL is in cell A1 of this workbook. Extract the base ID from that URL and use it to fetch the full schema. Write each table, field name, and field type as a row in the Schema tab.
When the BI lead wants the dictionary organized by table with section breaks
Flat rows work for filtering but the BI lead wants to read it top to bottom by table.
Fetch the schema of my Airtable 'Operations Hub' base. In the Schema tab, write the table name as a bold header row, then list all fields for that table below it. Insert a blank row between each table section.
When you want to compare the schemas of a staging and a production base
You have both bases and need to find differences before promoting a schema change.
Fetch the full schema of my Airtable base 'Operations Hub (Staging)' and write it into the Staging tab. Fetch the full schema of 'Operations Hub (Production)' and write it into the Production tab. In a Diff tab, list any field that exists in one base but not the other, and any field where the type differs between the two.
When you want the schema plus sample values from each field
The BI lead wants to see actual values, not just types, to decide on warehouse column widths.
Fetch the schema of my Airtable 'Operations Hub' base and write the data dictionary into the Schema tab. For each text and single-select field, pull up to 5 distinct sample values from the first 100 records in that table and add them as a comma-separated list in a Sample Values column.
The pattern: one prompt to the Airtable metadata API gives you what would take half a day of clicking through field editors.
Try It
Get the 7-day free trial of SheetXAI and open a blank workbook, then ask SheetXAI to pull the schema from any Airtable base you have access to. The Airtable integration is included in every SheetXAI plan. See also how to export filtered Airtable records into a workbook or the Airtable in Excel overview.