A comprehensive AI agent system designed for documentation-driven development with single feature focus and systematic execution. This system provides specialized agents that work independently while maintaining consistency through shared documentation standards.
Specializes in product management and requirements definition:
- PRD Creation: Craft comprehensive, actionable product requirements
- Feature Analysis: Evaluate and prioritize features based on impact and feasibility
- Strategic Planning: Assist in roadmap planning and market positioning
- Tech Stack Alignment: Ensure requirements align with ElysiaJS, React 19, SQLite, Prisma
- Backend Testing: Always specify comprehensive test coverage for backend features
Focuses on interface design with simplicity first and pattern consistency:
- Interface Design: Create clean, intuitive, and accessible user interfaces
- Pattern Application: Follow established design patterns and platform conventions
- Design System: Maintain consistent components and interactions
- Accessibility: Ensure WCAG compliance and inclusive design
- Responsive Design: Optimize for all screen sizes and devices
Implements documentation into functional code with single feature focus:
- Single Feature Development: Work on ONE feature at a time until complete
- Granular Steps: Break features into 30-60 minute implementable chunks
- Test-First Approach: Always create test files for backend functionality
- Context7 Integration: Use latest documentation for all tools and frameworks
- Quality Assurance: Ensure code meets documented standards
- ElysiaJS: Fast and type-safe web framework for Bun
- SQLite: Lightweight, serverless database
- Prisma: Type-safe database ORM and query builder
- React 19: Latest React with new features and improvements
- TanStack React Query: Powerful data fetching and state management
- React Router v7: Client-side routing and navigation
- Tailwind CSS: Utility-first CSS framework
- Tailwind Variants: Component variants and styling system
- Backend: Comprehensive test coverage required for all functionality
- Frontend: Optional testing for complex components
- Documentation-Driven: All decisions based on shared documentation
- Single Feature Focus: Work on ONE feature at a time until complete
- Granular Development: Break features into 30-60 minute steps
- Test Creation: Always create test files for backend features
- Context7 Research: Use latest documentation before implementation
- NEVER work on multiple features simultaneously
- NEVER skip test file creation for backend features
- NEVER run CLI commands (provide instructions instead)
- ALWAYS use context7 for latest documentation
- ALWAYS break features into granular steps
- ALWAYS complete one step before moving to next
- Critical:
/docs/implementation.md,/docs/bug_tracking.md,/docs/project_structure.md - Specification:
/docs/ui_ux_doc.md,/docs/api_documentation.md,/docs/testing_requirements.md - Reference:
/docs/coding_standards.md,/docs/architecture_decisions.md
Our system prioritizes documentation consistency over dynamic inter-agent communication. All agents synchronize through well-defined documentation structures, ensuring:
- Consistency: All agents work from the same information base
- Traceability: Every decision can be traced back to documented requirements
- Maintainability: Changes managed through documentation updates
- Predictability: Deterministic and reproducible agent behavior
Each agent follows structured workflow protocols that eliminate ambiguity:
- Pre-Task Protocol: Mandatory documentation review before any action
- Task Execution Protocol: Step-by-step procedures for completing work
- Post-Task Protocol: Documentation updates and quality verification
The system emphasizes familiar patterns over novel solutions:
- Proven Patterns: Prefer established solutions over innovative approaches
- Cognitive Load Reduction: Minimize mental effort required for understanding
- Consistency Over Creativity: Maintain predictable behaviors and outputs
Focus on one feature at a time with granular execution:
- Complete Implementation: Finish current feature entirely before starting next
- 30-60 Minute Steps: Break features into manageable, testable chunks
- Test-Driven: Create comprehensive test coverage for backend functionality
- Quality Gates: Validate each step before proceeding to next
- Deterministic behavior through explicit rules and documentation requirements
- Consistent output quality across different contexts
- Clear traceability when issues arise
- Eliminates unpredictable inter-agent interactions
- Standardizes interfaces and interaction patterns
- Centralizes knowledge in documentation
- Clear documentation with all decisions explicitly recorded
- Standardized processes across all agents
- Modular design allows independent agent updates
- Documentation structures grow systematically
- Established patterns apply to new domains
- Standards propagate to new agents effectively
- Setup Documentation: Create the required
/docs/structure - Configure Agents: Set up each agent with their respective system prompts
- Define Project Rules: Establish project-specific development standards
- Start Development: Begin with single feature development using granular steps
/docs/
├── implementation.md # Current tasks and progress
├── bug_tracking.md # Known issues and solutions
├── project_structure.md # File organization rules
├── ui_ux_doc.md # Design patterns and components
├── api_documentation.md # API specifications
├── testing_requirements.md # Testing standards
├── coding_standards.md # Code style and conventions
└── architecture_decisions.md # Technical architecture rationale
This system represents a paradigm shift from dynamic, emergent AI behaviors to systematic, predictable outcomes. By prioritizing documentation consistency, explicit behavioral rules, and structured workflows, we create more maintainable, scalable, and reliable AI systems.
The approach's emphasis on simplicity over complexity, consistency over creativity, and documentation over communication provides a solid foundation for building AI agent systems that serve human needs effectively while maintaining predictable, high-quality outcomes.
Built with the philosophy of documentation-driven development and systematic execution for reliable, maintainable AI agent systems.