Repolode
Submit
supabase-community

supabase-community/seed

Seed automatically fills a database with production-like dummy data generated from your schema, for local development and testing.

View on GitHub
Stars
788
Forks
37
Open issues
25
Last commit
last month

What is it?

Seed is a TypeScript tool that generates realistic dummy data for a database based on its schema. It determines column values automatically, so developers only need to specify the values they care about, and it adds built-in reference data such as country lists and currency codes by default.

It creates related records automatically, so you do not have to track foreign-key IDs across tables when defining data. Seed generates a type-safe TypeScript client from your database structure, letting you write seed data with the full TypeScript and Node.js ecosystem. Data generation uses Copycat and is deterministic, so the same inputs always produce the same output, which suits repeatable tests.

Optionally, an LLM (via OpenAI or Groq API keys) can generate example values for text columns. These examples are saved to a editable JSON file, and running a sync command regenerates them from updated column descriptions.

Key features

  • 01Automatically infers column values from the database schema.
  • 02Creates related records without manual ID tracking.
  • 03Generates a type-safe TypeScript client from your database structure.
  • 04Deterministic data generation for repeatable tests.
  • 05Optional LLM-generated example values for text columns via OpenAI or Groq.

Use cases

  • Populating a local development database with realistic data.
  • Creating consistent fixtures for automated tests.
  • Generating text examples for text columns with an LLM.

Quick start

  1. 1

    Try out Seed by running the init command

    npx @snaplet/seed init

Things to know

  • AI-generated examples require an OpenAI or Groq API key.
  • Contributing setup relies on Nix and devenv, which adds significant setup overhead.

In collections

Databases & SQL toolsTesting frameworks & test automation tools

Topics

#database#testing#seed-data#typescript#postgresql#test-data#developer-tools

FAQ

What is seed?
Seed automatically fills a database with production-like dummy data generated from your schema, for local development and testing. Seed is a TypeScript tool that generates realistic dummy data for a database based on its schema. It determines column values automatically, so developers only need to specify the values they care about, and it adds built-in reference data such as country lists and currency codes by default.
Is seed free to use? What license does it have?
seed is open source under the MIT license. Check the license terms before using it in commercial products.
Is seed still maintained?
Updated within 6 months. The last commit to seed was on Oct 1, 2026.
What can I use seed for?
Populating a local development database with realistic data. Creating consistent fixtures for automated tests. Generating text examples for text columns with an LLM.
How do I install seed?
Try out Seed by running the init command: npx @snaplet/seed init
Where can I get seed?
The source code and installation instructions are on GitHub: https://github.com/supabase-community/seed

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects