docs/unwind/layers/database/
├── index.md           # Overview, table count, links to sections
├── schema.md          # Current DDL for all tables
├── repositories.md    # Data access patterns, queries
└── jsonb-schemas.md   # Complex field type definitions (if any JSONB/JSON columns)

docs/unwind/layers/database/
├── index.md
├── users-domain.md    # users, user_settings, user_roles tables
├── orders-domain.md   # orders, order_items, shipments tables
└── ...

mkdir -p docs/unwind/layers/database/

# Database Layer

## Sections
- [Schema](schema.md) - _pending_
- [Repositories](repositories.md) - _pending_
- [JSONB Schemas](jsonb-schemas.md) - _pending_

## Summary
_Analysis in progress..._

# Database Layer

## Sections
- [Schema](schema.md) - 12 tables, 4 indexes
- [Repositories](repositories.md) - 8 repository classes
- [JSONB Schemas](jsonb-schemas.md) - 3 complex field types

## Migrations

**Location:** `src/db/migrations/`

Current schema state (result of all migrations) is documented in [schema.md](schema.md).

## Entity Relationships

```mermaid
erDiagram
    users ||--o{ orders : places
    orders ||--|{ order_items : contains
    order_items }|--|| products : references

### schema.md

```markdown
# Database Schema

## Tables (12 total)

### users [MUST]

```sql
CREATE TABLE users (
    id BIGINT PRIMARY KEY AUTO_INCREMENT,
    email VARCHAR(255) NOT NULL UNIQUE,
    password_hash VARCHAR(255) NOT NULL,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

CREATE INDEX idx_users_email ON users(email);

### repositories.md

```markdown
# Repositories

## UserRepository

[UserRepository.java](https://github.com/owner/repo/blob/main/src/repository/UserRepository.java)

```java
public interface UserRepository extends JpaRepository<User, Long> {
    Optional<User> findByEmail(String email);

    @Query("SELECT u FROM User u WHERE u.status = :status")
    List<User> findByStatus(@Param("status") UserStatus status);
}

## Additional Requirements

### Field-Level Documentation [MUST]

For EVERY table, document ALL of the following:
- Column name and database type (VARCHAR, INTEGER, JSONB, etc.)
- NOT NULL constraints
- DEFAULT values
- UNIQUE constraints
- Foreign key relationships with ON DELETE behavior (CASCADE, SET NULL, RESTRICT)

**Example:**
```markdown
### users table [MUST]

| Column | Type | Nullable | Default | Constraints |
|--------|------|----------|---------|-------------|
| id | SERIAL | NO | auto | PRIMARY KEY |
| email | VARCHAR(255) | NO | - | UNIQUE |
| organisation | INTEGER | NO | - | FK → organisation.id ON DELETE CASCADE |
| created_at | TIMESTAMP | NO | NOW() | - |

### calculationData (JSONB) [MUST]

**Source:** Inferred from `snapshot-operations.ts:180-195`

```typescript
{
  periodIntervals: number;
  intervalType: 'hour' | 'day' | 'week' | 'month';
  total: number;
  capexPercentage: number;  // 0-100
  totalCapex: number;
  totalOpex: number;
}

### Index Documentation [SHOULD]

Document ALL indexes with:
- Index name
- Columns covered
- Type (btree, gin, partial)
- Rationale (if apparent from naming or usage)

## Mandatory Tagging

**Every table, function, and schema must have a [MUST], [SHOULD], or [DON'T] tag in its heading.**

Default categorizations for database layer:
- **[MUST]**: All tables, core repository functions, JSONB schemas
- **[SHOULD]**: Audit/logging tables, test utilities, performance indexes
- **[DON'T]**: ORM-specific query patterns, migration-specific syntax

Example:
```markdown
### users [MUST]
### audit_logs [SHOULD]
### FindUserByEmail [MUST]
### GetTestDBPath [SHOULD]