app-design-generation
About
This Claude Skill generates comprehensive application design documents when triggered by phrases like "app design" or "design doc". It analyzes your codebase and user input to produce technology-agnostic overviews covering architecture, core features, and user experience. Use it during feature planning to quickly create high-level design documentation.
Documentation
App Design Generation
Overview
Generate comprehensive Application Design Document based on deep codebase analysis and user input. The document provides a high-level overview of the application's architecture, core features, user experience, and business logic while remaining technology-agnostic and focused on the "what" rather than the "how".
Quick Start
Generate app design document by analyzing codebase and user input.
Example:
- Detect trigger: User says "create design document" or "app design"
- Analyze codebase: Deep dive into structure, features, user flows
- Interactive Q&A: Ask about architecture, features, business logic
- Generate doc: Create comprehensive design document with architecture, features, UX
Result: Complete application design document for planning and onboarding.
When to Use
- User request contains "app design", "design doc", "create design document"
- Planning new features and need high-level design documentation
- Onboarding new team members who need application overview
- Documenting application architecture and business logic
Process
1. Initial Analysis
- Analyze project structure and existing codebase
- Review package.json for project name and dependencies
- Check for existing documentation in .cursor/rules/
- Identify key application features and patterns
- Think deeply about the application's purpose and architecture
2. Codebase Deep Dive
Analyze the codebase to understand:
- Application Structure: Main modules, features, and components
- User Flows: Authentication, navigation, key user journeys
- Data Models: Conceptual relationships and entities
- Business Logic: Core rules, workflows, and processes
- Integrations: External services and APIs
- Security Patterns: Authentication and authorization approaches
3. Interactive Q&A Session
CRITICAL: Ask project stage question FIRST, then 4-7 additional questions:
- Use lettered/numbered options for easy response
- Focus on business goals and user needs
- Gather context for proper documentation
Required Questions:
-
Project Stage Assessment (Ask First!):
- Pre-MVP / MVP / Production / Enterprise
-
Application Purpose & Users:
- What is the primary problem your application solves?
- Who are your target users and what are their main goals?
-
Unique Value Proposition:
- What makes your application unique compared to existing solutions?
-
User Roles & Permissions:
- What different types of users interact with your system?
-
Core User Journeys:
- What are the 2-3 most critical user flows?
-
Business Model & Growth:
- How does this application generate value?
-
Integration Ecosystem:
- What external services must you integrate with?
4. Update Project Configuration
Based on project stage response:
- Update
.cursor/rules/project-status.mdcwith current stage - Set appropriate DO/DON'T priorities for the stage
- Document stage-specific development guidelines
5. Generate Document
Create comprehensive app design document following standard structure:
- Introduction (overview, purpose, target audience, value proposition)
- Core Features (feature categories with purpose, functionalities, UX considerations)
- User Experience (personas, journeys, interface principles, accessibility)
- System Architecture (components, data flow, integrations, security)
- Business Logic (rules, processes, data models, workflows, validation)
- Future Considerations (enhancements, scalability, integrations, roadmap)
6. Save and Organize
- Create
.cursor/rules/directory if needed - Save as
app-design-document.mdc - Suggest next steps (tech stack doc, PRD, etc.)
Document Structure
The generated document must follow this high-level structure:
Introduction
- Application overview and purpose
- Target audience and user base
- Core value proposition
- Business context and goals
Core Features
- Feature Category 1: Purpose, functionalities, UX considerations
- Feature Category 2: Purpose, functionalities, UX considerations
- [Additional feature categories as needed]
User Experience
- User personas and roles
- Key user journeys and flows
- Interface design principles
- Accessibility and usability considerations
System Architecture
- High-level system components
- Data flow and relationships
- Integration points and external services
- Security and privacy approach
Business Logic
- Core business rules and processes
- Data models and relationships (conceptual)
- Workflow and state management
- Validation and business constraints
Future Considerations
- Planned enhancements and features
- Scalability considerations
- Potential integrations
- Long-term vision and roadmap
Writing Principles
DO:
- Business Focus: Describe WHAT the application does, not HOW
- User Value: Emphasize benefits and outcomes for users
- Clear Language: Write for non-technical stakeholders
- Visual Thinking: Use diagrams and flows where helpful
- Future Ready: Consider growth and evolution paths
DON'T:
-
Technical Details: No code snippets or implementation specifics
-
Technology Stack: Save for tech-stack.mdc document
-
Database Schemas: Keep data models conceptual
-
API Specifications: Focus on capabilities, not endpoints
-
Performance Metrics: Describe goals, not technical benchmarks
-
Suggest next steps (tech stack doc, PRD, etc.)
Troubleshooting
Common Issues:
-
Design document not generated
- Symptom: User requested but no document created
- Cause: Trigger keywords not detected or skill not invoked
- Fix: Check trigger keywords ("app design", "design doc"), invoke skill manually if needed
- Prevention: Verify trigger keywords in user request
-
Incomplete codebase analysis
- Symptom: Document missing key features or architecture details
- Cause: Didn't complete deep codebase analysis
- Fix: Complete all analysis steps: structure, features, flows, models, integrations
- Prevention: Always complete deep codebase analysis
-
Document not saved in correct location
- Symptom: Document created but not in
.cursor/rules/ - Cause: Wrong save location
- Fix: Save to
.cursor/rules/asapp-design-document.mdc - Prevention: Always use correct save location
- Symptom: Document created but not in
If issues persist:
- Verify trigger keywords were detected
- Check that deep codebase analysis was completed
- Ensure document saved to correct location
- Review process steps in skill
Output
- Format: Markdown (
.mdc) - Location:
.cursor/rules/ - Filename:
app-design-document.mdc
Integration with cc10x Orchestrator
This skill is invoked automatically by the PLAN workflow Phase 2 when:
- User request contains documentation keywords
- Missing design documentation is detected
- Documentation generation intent is identified
The skill executes BEFORE requirements intake, ensuring design documentation is available for planning.
Quick Install
/plugin add https://github.com/romiluz13/cc10x/tree/main/app-design-generationCopy and paste this command in Claude Code to install this skill
GitHub 仓库
Related Skills
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
llamaguard
OtherLlamaGuard is Meta's 7-8B parameter model for moderating LLM inputs and outputs across six safety categories like violence and hate speech. It offers 94-95% accuracy and can be deployed using vLLM, Hugging Face, or Amazon SageMaker. Use this skill to easily integrate content filtering and safety guardrails into your AI applications.
langchain
MetaLangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.
