The Scenario
A new backend engineer joined your team two weeks ago. She's been reading the codebase, looking at the Prisma schema file, and asking questions that make it clear the written schema and the actual database have drifted — migrations ran in production that never got committed back to the repo.
You've been handed the task of producing the ground-truth schema reference. There are 15 tables across 3 databases. You could run psql commands against each one, copy the describe output, and manually build a spreadsheet. Or you could ask SheetXAI to do it.
The bad version:
- SSH into the environment, run psql, connect to database one
- Execute describe for each of 15 tables, copy each block of output
- Paste into a spreadsheet, reformat manually — column types don't align, constraints are in the wrong format, foreign keys are buried in separate output
That's an hour of work before you've written a single word of actual documentation. And when the next migration runs, the sheet is immediately stale.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Google Sheet. It reads your spreadsheet, and through its built-in Prisma integration it can inspect your database schema and write the full structure — tables, columns, types, constraints, foreign keys — into your cells without a single psql command.
Open a blank sheet for your schema reference and try:
Inspect the schema of my Prisma database for project [ID] and create a sheet listing every table, column name, data type, nullable flag, and any constraints or foreign key references
What You Get
- One row per column across all tables
- Columns: table name, column name, data type, nullable (yes/no), primary key (yes/no), foreign key reference if present
- Tables grouped together so the sheet reads table by table
- Constraint notes written as plain English descriptions, not raw SQL syntax
What If the Data Is Not Quite Ready
I need three separate databases on three separate tabs
Pull the full schema from all 3 Prisma databases listed in column A and write one sheet tab per database, each with columns: table, column, type, nullable, primary key, foreign key
Some tables have indexes I want included in the reference
Inspect the schema of my Prisma database for project [ID], list every table and column with type, nullable, and primary key, then in a second pass add a column showing any indexes defined on each column
I want to flag columns that have no default and are not nullable — high-risk for inserts
Inspect the full schema of my Prisma database for project [ID], write all columns to this sheet, and in column G mark "insert-required" for any column that is NOT nullable and has no default value
Pull the schema, highlight foreign key relationships, and generate a plain-English description of each table in one shot
Inspect my Prisma database schema for project [ID], write the full column list to this sheet, in column H note any foreign key references as "references [table].[column]", and in column I write a one-sentence plain-English description of what each table appears to store based on its column names
The pattern: combine the schema pull, the annotation, and the plain-English interpretation in one prompt instead of running them as three separate steps.
Try It
Get the 7-day free trial of SheetXAI and open a blank sheet in any workbook where your team tracks database documentation, then ask it to inspect your Prisma schema and write the full reference. For related workflows, see the spoke on inventorying all Prisma projects or the hub overview.