ai-flow-dev 2.7.0 → 2.8.1

package/prompts/backend/flow-build-phase-2.md

@@ -1,573 +1,573 @@
-## PHASE 2: Data Architecture (15-20 min)
-
-> **Order for this phase:** 2.1 → 2.2 → 2.3 → 2.4 → 2.5 → 2.6 → 2.7
-
-> **📌 Scope-based behavior:**
-> - **MVP/Basic Scope:** Focus only on essential entities and basic relationships.
-> - **Production-Ready Scope:** In-depth modeling including indexes, constraints, and audit fields.
-
-### Objective
-
-Design the database model, entities, and relationships.
-
----
-
-## 🔍 Pre-Flight Check (Smart Skip Logic)
-
-> 📎 **Reference:** See [prompts/shared/smart-skip-preflight.md](../../.ai-flow/prompts/shared/smart-skip-preflight.md) for the complete smart skip logic.
-
-**Execute Pre-Flight Check for Phase 2:**
-
-- **Target File**: `docs/data-model.md`
-- **Phase Name**: "DATA ARCHITECTURE"
-- **Key Items**: Entities, relationships, data patterns, indexes
-- **Typical Gaps**: Missing entities, undocumented relationships, missing fields
-
-**Proceed with appropriate scenario based on audit data from `.ai-flow/cache/audit-data.json`**
-
----
-
-## Phase 2 Questions (Full Mode)
-
-**2.1 Database Type**
-
-```
-[If detected from Phase 0, show:]
-✅ Database Detected: [PostgreSQL/MySQL/MongoDB/etc.]
-✅ Version: [version if found]
-✅ ORM/Client: [Prisma/TypeORM/Sequelize/SQLAlchemy/etc.]
-
-Is this correct? (Y/N)
-If no, please provide correct database type.
-
-[If NOT detected, ask:]
-What type of database will you use? (Can select multiple)
-
-A) ⭐ PostgreSQL - Recommended for most backends (ACID, relational, JSON support)
-B) 🔥 MySQL/MariaDB - Popular, proven, wide ecosystem
-C) ⚡ MongoDB - Modern, NoSQL, flexible schema
-D) 🏆 Multi-database - PostgreSQL + Redis + S3, etc.
-E) Other: [specify]
-
-Why this choice?
-```
-
-**2.2 Core Data Entities**
-
-```
-[If detected from Phase 0, show:]
-✅ Entities Detected from Code:
-- [User] - [description if inferred from code]
-- [Product] - [description]
-- [Order] - [description]
-- [etc.]
-
-Are these correct? (Y/N)
-Do you need to add more entities? (Y/N)
-
-[If NOT detected OR user wants to add more, show:]
-Based on your system type (from Phase 1, question 1.5), here are common entities:
-
-🛒 E-commerce typical entities:
-1) User - System users with authentication
-2) Product - Items available for purchase
-3) Category - Product categorization
-4) Cart - Shopping cart items
-5) Order - Customer orders
-6) OrderItem - Individual items in an order
-7) Payment - Payment transactions
-8) Address - Shipping/billing addresses
-9) Review - Product reviews and ratings
-10) Inventory - Stock tracking
-
-📱 SaaS typical entities:
-1) User - System users
-2) Organization - Tenant/workspace
-3) Team - Groups within organizations
-4) Role - Access control roles
-5) Permission - Granular permissions
-6) Subscription - Billing plans
-7) Invoice - Payment records
-8) ApiKey - API access credentials
-9) AuditLog - Activity tracking
-
-📊 CRM typical entities:
-1) User - System users
-2) Contact - Customers/leads
-3) Company - Organizations
-4) Deal - Sales opportunities
-5) Activity - Calls, emails, meetings
-6) Task - To-do items
-7) Note - Free-form notes
-8) Document - Attachments
-
-🎮 Social typical entities:
-1) User - Platform users
-2) Profile - User profiles
-3) Post - Content/publications
-4) Comment - Post comments
-5) Like/Reaction - Engagement
-6) Follow - User connections
-7) Notification - User alerts
-H) Message - Direct messages
-I) Group - Communities
-
-→ Your selection (e.g., A, C, F): __
-
-OR list your custom entities:
-
-1.
-2.
-3.
-4.
-5.
-...
-
-(Include brief description for custom entities)
-```
-
-**2.3 Relationships**
-
-```
-Common relationship patterns (select what applies to your entities):
-
-⭐ One-to-Many (most common):
-A) User → Order (one user has many orders)
-B) User → Post (one user creates many posts)
-C) Organization → User (one org has many users)
-D) Category → Product (one category contains many products)
-E) Order → OrderItem (one order has many line items)
-F) Post → Comment (one post has many comments)
-G) Other: __
-
-→ Your selection (e.g., A, B, D): __
-
-⭐ Many-to-Many (via join table):
-A) Order ↔ Product (via OrderItem)
-B) User ↔ Role (via UserRole)
-E) Course ↔ Student (via Enrollment)
-F) Other: __
-
-→ Your selection (e.g., A, C, E): __
-
-⭐ One-to-One (less common):
-
-⭐ One-to-One (less common):
-C) Order → Payment (one order has one payment)
-D) Other: __
-
-→ Your selection (e.g., A, B): __
-
-⭐ Polymorphic (one entity relates to multiple types):
-C) Activity → (User | Organization | Deal) - activities linked to various objects
-D) Other: __
-
-→ Your selection (e.g., A, C): __
----
-Your specific relationships (list main ones):
----
-Your specific relationships (list main ones):
--
--
--
-
-(Format: EntityA → EntityB: Relationship type - description)
-```
-
-**2.4 Data Volume Estimates**
-
-```
-Estimated data volume (Year 1):
-
-- Total records: [Low (<10k) / Medium (10k-1M) / High (>1M)]
-- Growth rate: [Slow / Moderate / Fast]
-
-Data Complexity (Record Size):
-A) 📄 Low - Mostly text data (JSON, strings)
-B) 🖼️ Medium - Some images/documents (blobs, small files)
-C) 🎥 High - Heavy media/large files (video, audio, raw data)
-
-⭐ Standard for MVP:
-- Records: Low (<10k)
-- Growth: Moderate
-- Complexity: Low (mostly text)
-
-🏆 Standard for Production/Scale:
-- Records: High (>1M)
-- Growth: Fast
-- Complexity: Medium/High (includes media/files)
-```
-
-**2.5 Data Retention**
-
-```
-Data retention policies:
-
-A) ♾️ Keep forever - Never delete data
-B) 🗓️ Regulatory compliance - Specific retention period (e.g., 7 years)
-C) 🔄 Archival strategy - Archive old data after __ months
-D) 🗑️ Auto-deletion - Delete after __ days/months
-
-For each entity that has special retention needs, specify:
-```
-
-**2.6 Data Migration**
-
-```
-Is this a new system or replacing an existing one?
-
-A) 🆕 New system - No legacy data
-B) 🔄 Replacing existing - Need to migrate from [system name]
-C) 🔌 Integration - Syncing with existing system
-
-If migration needed:
-- Source system: __
-- Data volume to migrate: __
-- Migration strategy: [Big bang / Phased / Parallel run]
-```
-
-**2.7 Critical Data Patterns**
-
-```
-Select data patterns that apply:
-
-A) 🔐 Soft deletes - Keep deleted records with deleted_at flag
-B) 📝 Audit trail - Track who changed what and when
-C) 🕐 Temporal data - Track historical versions
-D) 🌍 Multi-tenancy - Data isolation per customer/organization
-E) 🎭 Polymorphic relationships - One entity relates to multiple types
-F) 🔗 Graph relationships - Complex many-to-many networks
-G) 📊 Aggregations/Materialized views - Pre-computed summaries
-H) 🗂️ Partitioning - Split large tables by date/region/etc.
-
-For each selected, provide brief detail:
-```
-
-**2.7.1 Soft Delete Configuration** (if A selected above)
-
-```
-How will you handle data deletion?
-
-Field for soft delete:
-A) ⭐ deleted_at (timestamp, null = active) - Recommended
-B) is_deleted (boolean)
-C) status field (e.g., status = 'deleted')
-D) Custom: __
-
-Entities with SOFT delete (keep record, mark as deleted):
-- Users ✅
-- Orders ✅
-- Products ✅
-- [List yours...]
-
-Entities with HARD delete (permanent removal):
-- Session tokens
-- Temporary files
-- Cart items after checkout
-- [List yours...]
-
-Permanent cleanup policy:
-A) ⭐ Purge soft-deleted after __ days (recommended: 90)
-B) Archive to cold storage after __ days
-C) Never delete (compliance requirement)
-
-Default query behavior:
-A) ⭐ Exclude deleted by default (add scope/filter)
-B) Include all, filter explicitly
-```
-
-**2.7.2 State Machines** (for entities with lifecycle states)
-
-```
-Do any entities have defined state lifecycles?
-
-A) ⭐ Yes - Define state machines
-B) No - Simple status fields without transitions
-
-If yes, define for each entity:
-
----
-Entity: Order (example)
-States: [draft, pending, confirmed, shipped, delivered, cancelled, refunded]
-
-Valid Transitions:
-- draft → pending (action: submit)
-- pending → confirmed (action: pay) [requires: payment_id]
-- pending → cancelled (action: cancel, or timeout: 24h)
-- confirmed → shipped (action: ship) [requires: tracking_number]
-- shipped → delivered (action: deliver)
-- confirmed → refunded (action: refund)
-- delivered → refunded (action: refund) [within: 30 days]
-
-Invalid Transitions (explicitly forbidden):
-- shipped → cancelled (cannot cancel after shipping)
-- delivered → cancelled
-
-Side Effects:
-- pending → confirmed: send confirmation email, reserve inventory
-- confirmed → cancelled: release inventory, refund payment
-- shipped → delivered: send delivery notification
----
-
-Your state machines:
-
-Entity: __
-States: __
-Transitions: __
-```
-
-**2.7.1 Domain-Driven Design Concepts** (Production-Ready and Enterprise only)
-
-```
-Will you use Domain-Driven Design (DDD) patterns?
-
-A) ⭐ Yes - Using DDD tactical patterns
-   - Aggregate Roots for transactional boundaries
-   - Value Objects for immutable data
-   - Domain Events for decoupling
-
-B) 🔥 Partial - Only Aggregate Roots
-   - Define aggregate roots for complex domains
-   - Keep entities grouped by aggregate
-
-C) No - Simple CRUD patterns
-   - Standard entity relationships
-   - No aggregate boundaries
-
-If using DDD (A or B):
-
-What are your Aggregate Roots?
-(Aggregate roots are the entry points to groups of related entities)
-
-Examples:
-- Order (with OrderItems, Shipping, Payment)
-- User (with Profile, Preferences, Settings)
-- Project (with Tasks, Members, Files)
-
-Your Aggregate Roots:
-1. __ (contains: __)
-2. __ (contains: __)
-3. __ (contains: __)
-
-Domain Events (things that happen in your domain):
-- UserRegistered
-- OrderPlaced
-- PaymentCompleted
-- etc.
-
-Your key domain events:
-1.
-2.
-3.
-```
-
-**2.7.4 Transaction Boundaries**
-
-```
-Which operations require database transactions?
-
-List operations that must be atomic (all-or-nothing):
-
-1. User Registration:
-   - Create User record
-   - Create Profile record
-   - Send welcome email (queue, not in transaction)
-   → Rollback if: User or Profile creation fails
-
-2. Order Creation:
-   - Create Order record
-   - Create OrderItems
-   - Reserve inventory
-   - Charge payment
-   → Rollback if: Any step fails
-
-3. [Your operations]:
-   - Step 1
-   - Step 2
-   → Rollback if: __
-
-Your transactional operations:
-1.
-2.
-3.
-```
-
-**2.8 Database Indexes**
-
-```
-What indexes will you need for performance optimization?
-
-Indexes are critical for query performance. Based on your entities and relationships, consider:
-
-Common indexes needed:
-A) Foreign keys (automatically indexed by most ORMs)
-B) Frequently queried columns (email, username, status)
-C) Columns used in WHERE clauses
-D) Columns used in JOIN conditions
-E) Columns used in ORDER BY clauses
-F) Composite indexes for multi-column queries
-
-→ Your selection (e.g., A, B, C, D): __
-
-Do you have specific query patterns that need optimization?
-
-Example:
-- User lookup by email: Index on users.email
-- Order search by date range: Index on orders.created_at
-- Product search by category and status: Composite index on (category_id, status)
-
-Your specific indexes:
-1.
-2.
-3.
-```
-
-**2.9 Transaction Management**
-
-```
-What transaction isolation level will you use?
-
-A) ⭐ READ COMMITTED - Recommended default (PostgreSQL, MySQL default)
-   - Prevents dirty reads
-   - Allows non-repeatable reads and phantom reads
-   - Good balance of consistency and performance
-
-B) READ UNCOMMITTED - Lowest isolation (rarely used)
-   - Allows dirty reads
-   - Fastest but least safe
-
-C) REPEATABLE READ - Higher isolation
-   - Prevents dirty reads and non-repeatable reads
-   - May have phantom reads
-   - Better consistency, slightly slower
-
-D) 🏆 SERIALIZABLE - Highest isolation (Enterprise)
-   - Prevents all concurrency issues
-   - Slowest but safest
-   - Use only when absolutely necessary
-
-Your choice: __
-
-Consistency model:
-A) ⭐ Strong consistency - All reads see latest writes (most backends)
-B) Eventual consistency - Acceptable delay for better performance (distributed systems)
-
-If eventual consistency:
-- Acceptable delay: __ seconds/minutes
-- Conflict resolution strategy: __
-```
-
-**2.10 Schema Migrations**
-
-```
-What migration tool will you use?
-
-A) ⭐ Prisma Migrate (if using Prisma)
-B) TypeORM Migrations (if using TypeORM)
-C) Alembic (Python/SQLAlchemy)
-D) Flyway (Java/Universal)
-E) Liquibase (Java/Universal)
-F) Django Migrations (Django)
-G) Laravel Migrations (Laravel)
-H) Rails Migrations (Ruby on Rails)
-I) Other: __
-
-Migration strategy:
-A) ⭐ Versioned migrations - Each change creates a new migration file
-B) Auto-migrations - Tool generates migrations automatically
-C) Manual SQL scripts - Write migrations manually
-
-Zero-downtime migrations:
-A) ⭐ Yes - Plan for zero-downtime migrations (Production-Ready/Enterprise)
-B) No - Accept maintenance windows (MVP)
-
-If zero-downtime:
-- Strategy: [Expand/Contract, Blue-Green, etc.]
-- Rollback plan: __
-```
-
-### Phase 2 Output
-
-```
-📋 PHASE 2 SUMMARY:
-
-Database: [type(s)]
-Core Entities: [list with descriptions]
-Relationships: [key relationships]
-Data Volume: [estimates]
-Retention: [policies]
-Migration: [strategy if applicable]
-Data Patterns: [selected patterns with brief details]
-Database Indexes: [list of indexes needed]
-Transaction Isolation: [level + consistency model]
-Schema Migrations: [tool + strategy + zero-downtime approach]
-
-Is this correct? (Yes/No)
-```
-
----
-
-### 📄 Generate Phase 2 Documents
-
-**Before starting generation:**
-
-```
-📖 Loading context from previous phases...
-✅ Re-reading project-brief.md
-```
-
-**Generate `docs/data-model.md` automatically:**
-
-- Use template: `.ai-flow/templates/docs/data-model.template.md`
-- Fill with all Phase 2 entity and relationship information
-- Include entity catalog, relationships, data patterns
-- Generate entity-relationship diagram (ER diagram) in mermaid format showing all entities and their relationships
-- Write to: `docs/data-model.md`
-
----
-
-#### 🎨 MERMAID ER DIAGRAM FORMAT - CRITICAL
-
-> 📎 **Reference:** See [prompts/shared/mermaid-guidelines.md](../../.ai-flow/prompts/shared/mermaid-guidelines.md) for ER diagram syntax, relationship notation, and common mistakes.
-
-## **Example ER Diagram:**
-
-```
-✅ Generated: docs/data-model.md
-
-Document has been created with all Phase 2 information.
-
-📝 Would you like to make any corrections before continuing?
-
-→ If yes: Edit docs/data-model.md and type "ready" when done. I'll re-read it.
-→ If no: Type "continue" to proceed to Phase 3.
-```
-
-**If user edits the file:**
-Execute `read_file('docs/data-model.md')` to refresh context before continuing.
-
----
-
-> ⚠️ **CRITICAL:** DO NOT generate README.md in Phase 2. README.md is ONLY generated in Phase 8 (step 8.5) after framework initialization.
-
----
-
-## 📝 Generated Documents
-
-After Phase 2, generate/update:
-- `docs/data-model.md` - Database schema and entity relationships
-
----
-
-**Next Phase:** Phase 3 - System Architecture (15-20 min)
-
-Read: `.ai-flow/prompts/backend/flow-build-phase-3.md`
-
----
-
-**Last Updated:** 2025-12-20
-**Version:** 2.1.8
-
----
-
-## PHASE 3: System Architecture (15-20 min)
+## PHASE 2: Data Architecture (15-20 min)
+
+> **Order for this phase:** 2.1 → 2.2 → 2.3 → 2.4 → 2.5 → 2.6 → 2.7
+
+> **📌 Scope-based behavior:**
+> - **MVP/Basic Scope:** Focus only on essential entities and basic relationships.
+> - **Production-Ready Scope:** In-depth modeling including indexes, constraints, and audit fields.
+
+### Objective
+
+Design the database model, entities, and relationships.
+
+---
+
+## 🔍 Pre-Flight Check (Smart Skip Logic)
+
+> 📎 **Reference:** See [prompts/shared/smart-skip-preflight.md](../../.ai-flow/prompts/shared/smart-skip-preflight.md) for the complete smart skip logic.
+
+**Execute Pre-Flight Check for Phase 2:**
+
+- **Target File**: `docs/data-model.md`
+- **Phase Name**: "DATA ARCHITECTURE"
+- **Key Items**: Entities, relationships, data patterns, indexes
+- **Typical Gaps**: Missing entities, undocumented relationships, missing fields
+
+**Proceed with appropriate scenario based on audit data from `.ai-flow/cache/audit-data.json`**
+
+---
+
+## Phase 2 Questions (Full Mode)
+
+**2.1 Database Type**
+
+```
+[If detected from Phase 0, show:]
+✅ Database Detected: [PostgreSQL/MySQL/MongoDB/etc.]
+✅ Version: [version if found]
+✅ ORM/Client: [Prisma/TypeORM/Sequelize/SQLAlchemy/etc.]
+
+Is this correct? (Y/N)
+If no, please provide correct database type.
+
+[If NOT detected, ask:]
+What type of database will you use? (Can select multiple)
+
+A) ⭐ PostgreSQL - Recommended for most backends (ACID, relational, JSON support)
+B) 🔥 MySQL/MariaDB - Popular, proven, wide ecosystem
+C) ⚡ MongoDB - Modern, NoSQL, flexible schema
+D) 🏆 Multi-database - PostgreSQL + Redis + S3, etc.
+E) Other: [specify]
+
+Why this choice?
+```
+
+**2.2 Core Data Entities**
+
+```
+[If detected from Phase 0, show:]
+✅ Entities Detected from Code:
+- [User] - [description if inferred from code]
+- [Product] - [description]
+- [Order] - [description]
+- [etc.]
+
+Are these correct? (Y/N)
+Do you need to add more entities? (Y/N)
+
+[If NOT detected OR user wants to add more, show:]
+Based on your system type (from Phase 1, question 1.5), here are common entities:
+
+🛒 E-commerce typical entities:
+1) User - System users with authentication
+2) Product - Items available for purchase
+3) Category - Product categorization
+4) Cart - Shopping cart items
+5) Order - Customer orders
+6) OrderItem - Individual items in an order
+7) Payment - Payment transactions
+8) Address - Shipping/billing addresses
+9) Review - Product reviews and ratings
+10) Inventory - Stock tracking
+
+📱 SaaS typical entities:
+1) User - System users
+2) Organization - Tenant/workspace
+3) Team - Groups within organizations
+4) Role - Access control roles
+5) Permission - Granular permissions
+6) Subscription - Billing plans
+7) Invoice - Payment records
+8) ApiKey - API access credentials
+9) AuditLog - Activity tracking
+
+📊 CRM typical entities:
+1) User - System users
+2) Contact - Customers/leads
+3) Company - Organizations
+4) Deal - Sales opportunities
+5) Activity - Calls, emails, meetings
+6) Task - To-do items
+7) Note - Free-form notes
+8) Document - Attachments
+
+🎮 Social typical entities:
+1) User - Platform users
+2) Profile - User profiles
+3) Post - Content/publications
+4) Comment - Post comments
+5) Like/Reaction - Engagement
+6) Follow - User connections
+7) Notification - User alerts
+H) Message - Direct messages
+I) Group - Communities
+
+→ Your selection (e.g., A, C, F): __
+
+OR list your custom entities:
+
+1.
+2.
+3.
+4.
+5.
+...
+
+(Include brief description for custom entities)
+```
+
+**2.3 Relationships**
+
+```
+Common relationship patterns (select what applies to your entities):
+
+⭐ One-to-Many (most common):
+A) User → Order (one user has many orders)
+B) User → Post (one user creates many posts)
+C) Organization → User (one org has many users)
+D) Category → Product (one category contains many products)
+E) Order → OrderItem (one order has many line items)
+F) Post → Comment (one post has many comments)
+G) Other: __
+
+→ Your selection (e.g., A, B, D): __
+
+⭐ Many-to-Many (via join table):
+A) Order ↔ Product (via OrderItem)
+B) User ↔ Role (via UserRole)
+E) Course ↔ Student (via Enrollment)
+F) Other: __
+
+→ Your selection (e.g., A, C, E): __
+
+⭐ One-to-One (less common):
+
+⭐ One-to-One (less common):
+C) Order → Payment (one order has one payment)
+D) Other: __
+
+→ Your selection (e.g., A, B): __
+
+⭐ Polymorphic (one entity relates to multiple types):
+C) Activity → (User | Organization | Deal) - activities linked to various objects
+D) Other: __
+
+→ Your selection (e.g., A, C): __
+---
+Your specific relationships (list main ones):
+---
+Your specific relationships (list main ones):
+-
+-
+-
+
+(Format: EntityA → EntityB: Relationship type - description)
+```
+
+**2.4 Data Volume Estimates**
+
+```
+Estimated data volume (Year 1):
+
+- Total records: [Low (<10k) / Medium (10k-1M) / High (>1M)]
+- Growth rate: [Slow / Moderate / Fast]
+
+Data Complexity (Record Size):
+A) 📄 Low - Mostly text data (JSON, strings)
+B) 🖼️ Medium - Some images/documents (blobs, small files)
+C) 🎥 High - Heavy media/large files (video, audio, raw data)
+
+⭐ Standard for MVP:
+- Records: Low (<10k)
+- Growth: Moderate
+- Complexity: Low (mostly text)
+
+🏆 Standard for Production/Scale:
+- Records: High (>1M)
+- Growth: Fast
+- Complexity: Medium/High (includes media/files)
+```
+
+**2.5 Data Retention**
+
+```
+Data retention policies:
+
+A) ♾️ Keep forever - Never delete data
+B) 🗓️ Regulatory compliance - Specific retention period (e.g., 7 years)
+C) 🔄 Archival strategy - Archive old data after __ months
+D) 🗑️ Auto-deletion - Delete after __ days/months
+
+For each entity that has special retention needs, specify:
+```
+
+**2.6 Data Migration**
+
+```
+Is this a new system or replacing an existing one?
+
+A) 🆕 New system - No legacy data
+B) 🔄 Replacing existing - Need to migrate from [system name]
+C) 🔌 Integration - Syncing with existing system
+
+If migration needed:
+- Source system: __
+- Data volume to migrate: __
+- Migration strategy: [Big bang / Phased / Parallel run]
+```
+
+**2.7 Critical Data Patterns**
+
+```
+Select data patterns that apply:
+
+A) 🔐 Soft deletes - Keep deleted records with deleted_at flag
+B) 📝 Audit trail - Track who changed what and when
+C) 🕐 Temporal data - Track historical versions
+D) 🌍 Multi-tenancy - Data isolation per customer/organization
+E) 🎭 Polymorphic relationships - One entity relates to multiple types
+F) 🔗 Graph relationships - Complex many-to-many networks
+G) 📊 Aggregations/Materialized views - Pre-computed summaries
+H) 🗂️ Partitioning - Split large tables by date/region/etc.
+
+For each selected, provide brief detail:
+```
+
+**2.7.1 Soft Delete Configuration** (if A selected above)
+
+```
+How will you handle data deletion?
+
+Field for soft delete:
+A) ⭐ deleted_at (timestamp, null = active) - Recommended
+B) is_deleted (boolean)
+C) status field (e.g., status = 'deleted')
+D) Custom: __
+
+Entities with SOFT delete (keep record, mark as deleted):
+- Users ✅
+- Orders ✅
+- Products ✅
+- [List yours...]
+
+Entities with HARD delete (permanent removal):
+- Session tokens
+- Temporary files
+- Cart items after checkout
+- [List yours...]
+
+Permanent cleanup policy:
+A) ⭐ Purge soft-deleted after __ days (recommended: 90)
+B) Archive to cold storage after __ days
+C) Never delete (compliance requirement)
+
+Default query behavior:
+A) ⭐ Exclude deleted by default (add scope/filter)
+B) Include all, filter explicitly
+```
+
+**2.7.2 State Machines** (for entities with lifecycle states)
+
+```
+Do any entities have defined state lifecycles?
+
+A) ⭐ Yes - Define state machines
+B) No - Simple status fields without transitions
+
+If yes, define for each entity:
+
+---
+Entity: Order (example)
+States: [draft, pending, confirmed, shipped, delivered, cancelled, refunded]
+
+Valid Transitions:
+- draft → pending (action: submit)
+- pending → confirmed (action: pay) [requires: payment_id]
+- pending → cancelled (action: cancel, or timeout: 24h)
+- confirmed → shipped (action: ship) [requires: tracking_number]
+- shipped → delivered (action: deliver)
+- confirmed → refunded (action: refund)
+- delivered → refunded (action: refund) [within: 30 days]
+
+Invalid Transitions (explicitly forbidden):
+- shipped → cancelled (cannot cancel after shipping)
+- delivered → cancelled
+
+Side Effects:
+- pending → confirmed: send confirmation email, reserve inventory
+- confirmed → cancelled: release inventory, refund payment
+- shipped → delivered: send delivery notification
+---
+
+Your state machines:
+
+Entity: __
+States: __
+Transitions: __
+```
+
+**2.7.1 Domain-Driven Design Concepts** (Production-Ready and Enterprise only)
+
+```
+Will you use Domain-Driven Design (DDD) patterns?
+
+A) ⭐ Yes - Using DDD tactical patterns
+   - Aggregate Roots for transactional boundaries
+   - Value Objects for immutable data
+   - Domain Events for decoupling
+
+B) 🔥 Partial - Only Aggregate Roots
+   - Define aggregate roots for complex domains
+   - Keep entities grouped by aggregate
+
+C) No - Simple CRUD patterns
+   - Standard entity relationships
+   - No aggregate boundaries
+
+If using DDD (A or B):
+
+What are your Aggregate Roots?
+(Aggregate roots are the entry points to groups of related entities)
+
+Examples:
+- Order (with OrderItems, Shipping, Payment)
+- User (with Profile, Preferences, Settings)
+- Project (with Tasks, Members, Files)
+
+Your Aggregate Roots:
+1. __ (contains: __)
+2. __ (contains: __)
+3. __ (contains: __)
+
+Domain Events (things that happen in your domain):
+- UserRegistered
+- OrderPlaced
+- PaymentCompleted
+- etc.
+
+Your key domain events:
+1.
+2.
+3.
+```
+
+**2.7.4 Transaction Boundaries**
+
+```
+Which operations require database transactions?
+
+List operations that must be atomic (all-or-nothing):
+
+1. User Registration:
+   - Create User record
+   - Create Profile record
+   - Send welcome email (queue, not in transaction)
+   → Rollback if: User or Profile creation fails
+
+2. Order Creation:
+   - Create Order record
+   - Create OrderItems
+   - Reserve inventory
+   - Charge payment
+   → Rollback if: Any step fails
+
+3. [Your operations]:
+   - Step 1
+   - Step 2
+   → Rollback if: __
+
+Your transactional operations:
+1.
+2.
+3.
+```
+
+**2.8 Database Indexes**
+
+```
+What indexes will you need for performance optimization?
+
+Indexes are critical for query performance. Based on your entities and relationships, consider:
+
+Common indexes needed:
+A) Foreign keys (automatically indexed by most ORMs)
+B) Frequently queried columns (email, username, status)
+C) Columns used in WHERE clauses
+D) Columns used in JOIN conditions
+E) Columns used in ORDER BY clauses
+F) Composite indexes for multi-column queries
+
+→ Your selection (e.g., A, B, C, D): __
+
+Do you have specific query patterns that need optimization?
+
+Example:
+- User lookup by email: Index on users.email
+- Order search by date range: Index on orders.created_at
+- Product search by category and status: Composite index on (category_id, status)
+
+Your specific indexes:
+1.
+2.
+3.
+```
+
+**2.9 Transaction Management**
+
+```
+What transaction isolation level will you use?
+
+A) ⭐ READ COMMITTED - Recommended default (PostgreSQL, MySQL default)
+   - Prevents dirty reads
+   - Allows non-repeatable reads and phantom reads
+   - Good balance of consistency and performance
+
+B) READ UNCOMMITTED - Lowest isolation (rarely used)
+   - Allows dirty reads
+   - Fastest but least safe
+
+C) REPEATABLE READ - Higher isolation
+   - Prevents dirty reads and non-repeatable reads
+   - May have phantom reads
+   - Better consistency, slightly slower
+
+D) 🏆 SERIALIZABLE - Highest isolation (Enterprise)
+   - Prevents all concurrency issues
+   - Slowest but safest
+   - Use only when absolutely necessary
+
+Your choice: __
+
+Consistency model:
+A) ⭐ Strong consistency - All reads see latest writes (most backends)
+B) Eventual consistency - Acceptable delay for better performance (distributed systems)
+
+If eventual consistency:
+- Acceptable delay: __ seconds/minutes
+- Conflict resolution strategy: __
+```
+
+**2.10 Schema Migrations**
+
+```
+What migration tool will you use?
+
+A) ⭐ Prisma Migrate (if using Prisma)
+B) TypeORM Migrations (if using TypeORM)
+C) Alembic (Python/SQLAlchemy)
+D) Flyway (Java/Universal)
+E) Liquibase (Java/Universal)
+F) Django Migrations (Django)
+G) Laravel Migrations (Laravel)
+H) Rails Migrations (Ruby on Rails)
+I) Other: __
+
+Migration strategy:
+A) ⭐ Versioned migrations - Each change creates a new migration file
+B) Auto-migrations - Tool generates migrations automatically
+C) Manual SQL scripts - Write migrations manually
+
+Zero-downtime migrations:
+A) ⭐ Yes - Plan for zero-downtime migrations (Production-Ready/Enterprise)
+B) No - Accept maintenance windows (MVP)
+
+If zero-downtime:
+- Strategy: [Expand/Contract, Blue-Green, etc.]
+- Rollback plan: __
+```
+
+### Phase 2 Output
+
+```
+📋 PHASE 2 SUMMARY:
+
+Database: [type(s)]
+Core Entities: [list with descriptions]
+Relationships: [key relationships]
+Data Volume: [estimates]
+Retention: [policies]
+Migration: [strategy if applicable]
+Data Patterns: [selected patterns with brief details]
+Database Indexes: [list of indexes needed]
+Transaction Isolation: [level + consistency model]
+Schema Migrations: [tool + strategy + zero-downtime approach]
+
+Is this correct? (Yes/No)
+```
+
+---
+
+### 📄 Generate Phase 2 Documents
+
+**Before starting generation:**
+
+```
+📖 Loading context from previous phases...
+✅ Re-reading project-brief.md
+```
+
+**Generate `docs/data-model.md` automatically:**
+
+- Use template: `.ai-flow/templates/docs/data-model.template.md`
+- Fill with all Phase 2 entity and relationship information
+- Include entity catalog, relationships, data patterns
+- Generate entity-relationship diagram (ER diagram) in mermaid format showing all entities and their relationships
+- Write to: `docs/data-model.md`
+
+---
+
+#### 🎨 MERMAID ER DIAGRAM FORMAT - CRITICAL
+
+> 📎 **Reference:** See [prompts/shared/mermaid-guidelines.md](../../.ai-flow/prompts/shared/mermaid-guidelines.md) for ER diagram syntax, relationship notation, and common mistakes.
+
+## **Example ER Diagram:**
+
+```
+✅ Generated: docs/data-model.md
+
+Document has been created with all Phase 2 information.
+
+📝 Would you like to make any corrections before continuing?
+
+→ If yes: Edit docs/data-model.md and type "ready" when done. I'll re-read it.
+→ If no: Type "continue" to proceed to Phase 3.
+```
+
+**If user edits the file:**
+Execute `read_file('docs/data-model.md')` to refresh context before continuing.
+
+---
+
+> ⚠️ **CRITICAL:** DO NOT generate README.md in Phase 2. README.md is ONLY generated in Phase 8 (step 8.5) after framework initialization.
+
+---
+
+## 📝 Generated Documents
+
+After Phase 2, generate/update:
+- `docs/data-model.md` - Database schema and entity relationships
+
+---
+
+**Next Phase:** Phase 3 - System Architecture (15-20 min)
+
+Read: `.ai-flow/prompts/backend/flow-build-phase-3.md`
+
+---
+
+**Last Updated:** 2025-12-20
+**Version:** 2.1.8
+
+---
+
+## PHASE 3: System Architecture (15-20 min)