@autobe/agent/src/orchestrate/analyze/structures/IAutoBeAnalyzeScenarioApplication.ts

AI backend server code generator

import { AutoBeAnalyzeActor, CamelCasePattern } from "@autobe/interface";
import { AutoBeAnalyzeFile } from "@autobe/interface/src/histories/contents/AutoBeAnalyzeFile";
import { tags } from "typia";

export interface IAutoBeAnalyzeScenarioApplication {
  /**
   * Composes project structure with actors and documentation files.
   *
   * Determines the list of user actors and documents to generate based on
   * requirements. If requirements are incomplete, returns empty arrays.
   *
   * @param input - Project prefix, actors, and file list
   */
  compose(input: IAutoBeAnalyzeScenarioApplication.IProps): void;
}

export namespace IAutoBeAnalyzeScenarioApplication {
  export interface IProps {
    /** Reason for the analysis and composition of the project structure. */
    reason: string;

    /**
     * Prefix for file names and variable names. This will be used for
     * organizing documentation files.
     *
     * DO: Use camelCase naming convention.
     */
    prefix: string & CamelCasePattern;

    /**
     * Actors to be assigned for the project.
     *
     * Each actor has:
     *
     * - `name`: Actor identifier (camelCase)
     * - `kind`: "guest" | "member" | "admin"
     * - `description`: Actor's permissions and capabilities
     */
    actors: AutoBeAnalyzeActor[];

    /**
     * Language for document content. When specified by the user, this takes
     * precedence over the locale setting for determining document language.
     */
    language?: string;

    /**
     * If the user has requested a specific number of pages, enter that number.
     * Otherwise, provide an appropriate number of documents needed to meet the
     * user's requirements. This number must always match the length of the
     * files property, must be greater than 1, and must include the table of
     * contents. For example, if the user requests 3 pages, the total should be
     * 4, including the table of contents.
     */
    page: number;

    /**
     * Array of document metadata objects defining files to be generated.
     *
     * Each array element is an AutoBeAnalyzeFile.Scenario object containing:
     * - filename: The output file name (e.g., "01-service-overview.md")
     * - reason: Why this document is being created
     * - documentType, outline, constraints, etc.: Metadata guiding content generation
     *
     * These documents represent business-focused planning documentation:
     * - Business requirements and functional specifications in natural language
     * - User journey mapping and use case scenarios
     * - Business rules and workflow definitions
     * - Service overview and business model description
     * - User actors and permission requirements (described in natural language)
     * - Business logic and validation rules
     * - DO NOT: Include database schemas, ERD, or API specifications
     * - DO: Write all requirements in natural language for clarity
     *
     * Generate metadata objects based on actual requirements gathered from conversation.
     * Do not create unnecessary documentation - only generate what is needed to
     * properly define the business requirements and system specifications.
     *
     * # Array Length Rules
     *
     * The array length must match the user's requested page count plus one for ToC.
     * For example: user requests 3 pages → generate 4 objects (1 ToC + 3 content).
     * If user does not specify a number, generate sufficient objects to adequately
     * document the service (typically 11+ objects including ToC).
     */
    files: Array<AutoBeAnalyzeFile.Scenario> & tags.MinItems<1>;
  }
}