Domain Driven Design

What it does

This skill helps AI agents analyze requirements and generate domain models, entities, and relationships using Domain-Driven Design (DDD) principles for software design. It focuses on aligning code with a deep understanding of the business domain by establishing a shared language between developers and domain experts. The core principle is that the model is the code, and vice versa, ensuring complexity is manageable and requirements are captured precisely.

When to use it

• When designing software where a close alignment with business processes is critical.
• During code reviews to ensure adherence to DDD principles, particularly regarding naming conventions.
• To evaluate existing domain models and suggest improvements based on DDD guidelines (scoring them 0-10).

Key capabilities

• Ubiquitous Language Enforcement: Helps identify and correct instances where technical jargon replaces domain terminology in code.
• Naming Guidance: Provides recommendations for class, method, event, and module names that reflect the business domain.
• Domain Model Scoring: Evaluates existing models on a scale of 0-10 based on DDD principles, suggesting improvements to reach a score of 10/10.
• Bounded Context Awareness: Recognizes that terms can have different meanings in different contexts and helps manage those differences.

Example prompts

• "Analyze this requirement: 'Customers should be able to submit orders.' Suggest class names reflecting the domain."
• "Review this code snippet for adherence to DDD principles, focusing on naming conventions. Provide a score (0-10) and suggestions for improvement."
• "What are some examples of how the term ‘Customer’ might have different meanings in different bounded contexts?"

Tips & gotchas

• The skill relies heavily on collaboration with domain experts to establish an accurate ubiquitous language.
• Naming difficulty can be a signal that the model is incorrect, prompting further investigation and refinement.
• Be aware of potential differences in terminology across bounded contexts—this is expected and needs careful management.