ddd-repository

One-line summary

A repository hides persistence behind the illusion of an in-memory collection of aggregates — domain code does not know whether storage is SQL, key-value, or in-memory.

When to use this skill

• Loading or saving aggregates by identity.
• Replacing ad-hoc DAOs that leak persistence types into the domain.
• Defining a small set of domain-meaningful queries (findActiveByCustomer, not findByStatusEqualsAndCustomerIdEquals).

When NOT to use this skill

• Value objects — they have no identity to look up by.
• Cross-aggregate queries for display purposes — build a read model, not a repository method.
• Reporting / analytics — those want a query layer optimized for reads, not a domain repository.

Core content

A repository serves one aggregate root. Its interface lives in the domain layer; its implementation lives in the infrastructure layer. Domain code depends only on the interface; infrastructure code wires the implementation.