import { AutoBeEventBase } from "./base/AutoBeEventBase"; /** * Event fired when the Database agent begins the database design process. * * This event marks the initiation of the sophisticated three-tier compiler * infrastructure that transforms business requirements into validated database * architectures through AST manipulation. The Database agent start represents * the beginning of the foundational data layer development that will support * all subsequent application functionality. * * The database design process that begins with this event will proceed through * component organization, schema creation, validation, and compilation to * produce production-ready Prisma schemas that maintain perfect semantic * integrity and syntactic correctness while accurately reflecting business * requirements. * * @author Samchon */ export interface AutoBeDatabaseStartEvent extends AutoBeEventBase<"databaseStart"> { /** * Reason why the Database agent was activated through function calling. * * Explains the specific circumstances that triggered the AI chatbot to invoke * the Database agent via function calling. This could include reasons such as * initial database design after requirements analysis completion, updating * database schemas due to requirement changes, regenerating data models to * reflect modified business logic, or creating additional database structures * for new functionality. * * Understanding the activation reason provides context for the database * design scope and helps stakeholders understand whether this represents * initial development, schema refinement, or expansion of existing data * architecture. */ reason: string; /** * Iteration number of the requirements analysis this database design is being * started for. * * Indicates which version of the requirements analysis this database design * will reflect. This step number ensures that the Database agent works with * the current requirements and helps track the evolution of database schemas * as business requirements and data needs change. * * The step value enables proper synchronization between database design * activities and the underlying requirements, ensuring that the data * architecture remains aligned with the current project scope and business * objectives throughout the iterative development process. */ step: number; }