# aerostack store

> Generate database schema changes from natural language and seed your database with realistic test data using AI-powered CLI commands.

AI-powered data store management — generate schema changes from natural language and seed databases with realistic data.

## Generate schema changes

```bash
aerostack store schema [intent]
```

Describe what you want in plain English. The agent analyzes your current schema and generates the SQL migration.

```bash
# Add a table
aerostack store schema "add a products table with name, price, and inventory"

# Modify existing table
aerostack store schema "add a soft delete column to the users table"

# Complex relationships
aerostack store schema "add order items linked to orders and products with quantity and unit price"
```

The command outputs a migration file ready to apply. Review it before running `aerostack db migrate apply`.

---

## Seed your database

```bash
aerostack store seed [flags]
```

| Flag | Description |
|------|-------------|
| `--smart` | Use AI to generate realistic, contextually appropriate seed data |

```bash
# Basic seed (uses existing seed files)
aerostack store seed

# AI-generated realistic seed data
aerostack store seed --smart
```

With `--smart`, the AI generates data that makes sense for your domain — names, emails, product names, prices, etc. — rather than generic placeholder values.

`aerostack store` uses the Project Knowledge Graph. Run `aerostack index` first for the best results.

## Example workflow

```bash
# 1. Describe what you need
aerostack store schema "add a blog posts table with title, body, slug, published status, and author"

# 2. Review the generated migration
cat migrations/0005_add_blog_posts.sql

# 3. Apply it
aerostack db migrate apply

# 4. Regenerate types
aerostack generate types

# 5. Seed with realistic data
aerostack store seed --smart
```