Initializes a new AutoBeAgent instance with the specified configuration.
Creates and configures the agent with AI vendor settings, behavioral context (locale/timezone), and compilation infrastructure. The agent can optionally resume from previous conversation histories to continue development sessions or build upon existing work.
The constructor sets up the internal MicroAgentica engine, initializes the development state from provided histories, and establishes the event dispatch system for real-time progress notifications. The agent becomes ready for conversation-driven development immediately after construction.
Configuration properties including AI vendor settings, behavioral context, compilation tools, and optional conversation histories for session continuation
Engages in conversation with the agent to drive the vibe coding process.
Accepts user input in multiple formats including simple text strings, single multimodal content items, or arrays of content supporting text, images, file uploads, and audio input. The conversation serves as the primary interface for expressing requirements, providing feedback, and guiding the development process through natural language interaction.
The agent analyzes the conversation context to determine appropriate actions, potentially activating specialized agents (Analyze, Prisma, Interface, Test, Realize) through function calling based on user needs. Real-time progress events are fired through registered listeners while the conversation processes.
Returns all history records generated during this conversation turn, including user messages, assistant responses, and any development activities triggered by the interaction. This enables clients to track both conversational flow and development progress.
User input as text, single content item, or multimodal array
Promise resolving to array of history records from this conversation
Retrieves all generated files from the current development session.
Transforms the complete conversation-driven development process into a comprehensive collection of deployable artifacts, including requirements documentation, database schemas, API specifications, NestJS implementation code, and test suites. The generated files represent a fully functional backend application ready for immediate deployment or further customization.
The method produces a meticulously organized project structure that reflects professional software development standards. Requirements analysis documents capture and formalize your conversational input into structured technical specifications, providing clear traceability from user intent to final implementation. Database artifacts include Prisma schemas with precise type definitions, relationships, and constraints, along with migration files for proper database initialization and evolution.
The API layer emerges through comprehensive OpenAPI specifications documenting every endpoint, request format, response structure, and error condition. Generated NestJS controllers, DTOs, and service classes implement these specifications with TypeScript's strong typing system providing compile-time safety. Quality assurance is embedded throughout with complete test suites covering both unit and end-to-end scenarios.
The database configuration specified through the dbms option
fundamentally shapes the entire generated codebase. PostgreSQL
configuration produces production-ready code with robust connection pooling
and enterprise-grade optimizations, while SQLite generates lightweight code
perfect for local development and rapid prototyping without external
dependencies.
All artifacts maintain perfect consistency across the chosen database system, from Prisma configurations and connection strings to Docker compose files and environment templates. This deep integration ensures immediate deployment compatibility without manual configuration adjustments.
Optionaloptions: Partial<IAutoBeGetFilesOptions>Configuration specifying the target database management system and other code generation preferences that influence the structure and characteristics of the generated project files
Promise resolving to key-value pairs mapping logical file paths to complete file contents for all generated development artifacts, ready for immediate file system operations, build integration, or deployment workflows
Retrieves the complete conversation and development history.
Returns the chronologically ordered record of all events from the current session including user messages, assistant responses, development phase activities, progress events, and completion notifications. This comprehensive history enables conversation replay, development process analysis, and understanding of how requirements evolved into working software.
The history provides complete transparency into the vibe coding process, showing both conversational interactions and behind-the-scenes development activities. This information is valuable for debugging, process improvement, and educational purposes to understand the agent's decision-making process.
Chronologically ordered array of all history records including messages, events, and development activities
Retrieves comprehensive AI token usage statistics for the current session.
Returns detailed breakdown of token consumption across all specialized agents and processing phases, enabling cost monitoring, performance analysis, and optimization of AI resource utilization. Statistics include aggregate totals and component-specific breakdowns with input/output categorization, caching analysis, and reasoning token tracking.
Token usage data is essential for understanding the computational costs of different development phases and optimizing AI efficiency. The breakdown helps identify which agents or operations consume the most resources, enabling targeted optimization efforts while maintaining development quality.
Comprehensive token usage statistics with detailed breakdowns by agent, operation type, and consumption category
Unregisters a previously registered event listener.
Removes the specified event listener from the agent's notification system, stopping further event notifications for that particular listener function. This is useful for cleanup, dynamic listener management, or when components no longer need to receive specific event notifications.
The listener function reference must exactly match the function that was originally registered with on for successful removal. If no matching listener is found, the operation has no effect.
Event type the listener was registered for
The exact listener function reference to remove
The agent instance for method chaining
Registers an event listener for specific development phase events.
Enables client applications to receive real-time notifications about conversation flow, development progress, and completion events throughout the vibe coding pipeline. Event listeners provide visibility into agent activities and enable responsive user interfaces that can display progress, handle artifacts, and provide feedback.
The type-safe event system ensures that listeners receive properly typed events corresponding to their registration type, enabling robust event handling without runtime type issues. Multiple listeners can be registered for the same event type to support complex notification requirements.
Event type to listen for (e.g., "analyzeComplete", "prismaStart")
Callback function that receives the typed event when fired
The agent instance for method chaining
Main agent class that orchestrates the entire vibe coding pipeline through conversation-driven development.
The AutoBeAgent serves as the central coordinator for the waterfall-based development process with spiral model iterative improvements. It manages the five specialized agents (Analyze, Prisma, Interface, Test, Realize) that transform user conversations into complete working applications through a sophisticated AST-based compilation infrastructure.
The agent operates through natural language conversation, supporting multimodal input including text, images, files, and audio. It maintains conversation history, tracks development progress through real-time events, and provides access to all generated artifacts including requirements documentation, database schemas, API specifications, test suites, and implementation code.
The vibe coding approach eliminates traditional development barriers by enabling users to express requirements naturally while the agent handles all technical implementation details through validated AST transformations and continuous quality assurance feedback loops.
Author
Samchon