Project Designer

Overview

Transform initial project concepts into comprehensive, well-architected designs through systematic analysis, feature enhancement, and technology recommendations. This skill guides you through structured ideation, architectural decision-making, and strategic planning for AI/ML agents, web applications, and mobile projects.

Workflow

Phase 1: Concept Understanding

Objective: Thoroughly understand the user's core idea, goals, and constraints.

Actions:

1. •

   Extract Core Information:

   • •Project type (AI/ML agent, web app, mobile app)
   • •Primary problem being solved
   • •Target users or personas
   • •Key constraints (budget, timeline, team size, technical expertise)
2. •

   Identify Gaps:

   • •Missing requirements or specifications
   • •Unclear objectives or success criteria
   • •Undefined user needs
3. •

   Ask Clarifying Questions: Use the AskUserQuestion tool when:

   • •Project scope is ambiguous
   • •Multiple architectural approaches are viable
   • •Technical requirements are unclear
   • •User needs are not fully specified

Phase 2: Architecture and Technology Guidance

Objective: Recommend appropriate architectural patterns and technology stacks based on project requirements.

Reference Materials: Load based on project type and needs:

• •references/ai-ml-patterns.md: For AI/ML agents, RAG systems, conversational AI, ML pipelines
• •references/web-app-patterns.md: For web/mobile applications, API design, full-stack systems
• •references/tech-stack-matrix.md: For technology selection across all project types

Actions:

1. •Identify project type and match to established architectural patterns
2. •Recommend technology stack based on team expertise, budget, scalability needs
3. •Explain architectural decisions with clear rationale and trade-offs

Phase 3: Feature Discovery and Enhancement

Objective: Expand the initial concept with comprehensive, prioritized features.

Reference Material: references/feature-frameworks.md

Actions:

1. •Apply feature discovery methods (Jobs-to-be-Done, User Story Mapping, Feature Analogies)
2. •Categorize features: Must-Have (MVP), Should-Have (Post-MVP), Could-Have (Future)
3. •Apply prioritization framework (RICE, ICE, Kano model)
4. •Consider enhancement dimensions: capability, UX, integration, intelligence

Phase 4: Structured Documentation

Objective: Produce comprehensive project documentation using standardized templates.

Asset Templates: Use appropriate templates based on user needs:

• •assets/templates/project-brief.md: Complete project specification
• •assets/templates/architecture-blueprint.md: Detailed technical architecture
• •assets/templates/feature-roadmap.md: Prioritized feature timeline

Phase 5: Iterative Refinement

Objective: Refine the design based on user feedback and emerging requirements.

Actions:

1. •Present recommendations clearly with critical decisions highlighted
2. •Respond to user feedback and adjust recommendations
3. •Provide detailed explanations of recommendations
4. •Start high-level, drill down into specifics as needed

Usage Patterns

Pattern 1: New Project Ideation

User presents initial idea → Understand requirements → Recommend architecture/stack → Enhance features → Generate project-brief.md

Pattern 2: Existing Project Enhancement

User has working system → Understand current state → Apply feature frameworks → Prioritize enhancements → Generate feature-roadmap.md

Pattern 3: Technology Stack Consultation

User needs tech guidance → Clarify requirements → Consult tech-stack-matrix.md → Recommend stack with rationale

Pattern 4: Architecture Design

User needs technical design → Gather requirements → Recommend pattern → Define components → Generate architecture-blueprint.md

Best Practices

Communication:

• •Follow workflow phases sequentially
• •Use progressive disclosure (high-level first, details when needed)
• •Provide concrete, implementable recommendations

Technical Recommendations:

• •Balance pragmatism with best practices
• •Consider team capabilities and constraints
• •Provide primary recommendation + alternatives with trade-offs

Feature Design:

• •Ground features in user problems and needs
• •Define clear MVP scope
• •Use quantitative prioritization frameworks

Reference Material Usage

Load references/ai-ml-patterns.md when: AI agents, RAG systems, ML pipelines, NLP/CV applications

Load references/web-app-patterns.md when: Web/mobile apps, API design, authentication, database patterns

Load references/tech-stack-matrix.md when: Technology selection, team/budget considerations, scalability needs

Load references/feature-frameworks.md when: Feature ideation, prioritization, user-centric design, roadmap planning

Anti-Patterns to Avoid

Don't:

• •Suggest over-engineered solutions for simple problems
• •Ignore constraints (budget, timeline, expertise)
• •Skip Phase 1 understanding to rush to recommendations
• •Load all references unnecessarily
• •Make assumptions without clarifying

Do:

• •Ask clarifying questions when unclear
• •Tailor recommendations to specific context
• •Provide rationale for recommendations
• •Use reference materials to support recommendations
• •Balance comprehensiveness with actionability