sf-agent-framework/sf-core/tasks/data-model-design.md

AI Agent Orchestration Framework for Salesforce Development - Two-phase architecture with 70% context reduction

# Data Model Design

## Purpose

Design scalable and efficient data models in Salesforce that support business
requirements while optimizing for platform performance and limits.

## Instructions

1. **Requirements Analysis**
   - Gather business entity requirements
   - Document data relationships
   - Identify transaction volumes
   - Define security requirements
   - Analyze reporting needs

2. **Object Model Design**
   - Leverage standard objects where possible
   - Design custom objects strategically
   - Plan object relationships (lookup vs master-detail)
   - Consider object limits and allocations
   - Design for reporting requirements

3. **Field Design and Optimization**
   - Define field types appropriately
   - Plan formula field usage
   - Design roll-up summaries efficiently
   - Implement validation rules
   - Consider field limits per object

4. **Relationship Architecture**
   - Design parent-child hierarchies
   - Plan many-to-many relationships
   - Optimize lookup relationships
   - Consider relationship limits
   - Design for query performance

5. **Large Data Volume Considerations**
   - Implement skinny tables where needed
   - Design efficient indexes
   - Plan data archival strategy
   - Optimize for query selectivity
   - Consider platform limits

6. **Security and Sharing Model**
   - Design with security in mind
   - Plan organization-wide defaults
   - Consider sharing implications
   - Design for field-level security
   - Optimize sharing calculations

## Input Requirements

- Business requirements document
- Entity relationship diagrams
- Data volume projections
- Security requirements
- Integration specifications
- Reporting requirements

## Output Format

- Data Model Design Document containing:
  - Conceptual data model
  - Logical data model
  - Physical implementation design
  - Object relationship diagram
  - Field specifications
  - Security model design
  - Performance considerations

## Design Principles

- **Normalization**: Balance with platform considerations
- **Scalability**: Design for growth
- **Performance**: Optimize for queries
- **Maintainability**: Keep it simple
- **Security**: Build in from start
- **Flexibility**: Allow for changes

## Platform Considerations

- Custom object limits
- Field limits per object
- Relationship limits
- Roll-up summary limits
- Sharing rule limits
- Data storage limits
- API query limits

## Best Practices

- Start with standard objects
- Denormalize thoughtfully
- Index strategically
- Avoid circular dependencies
- Document all decisions
- Test with realistic volumes
- Plan for data archival