Advanced: Pipeline and Internal Mechanics

This chapter provides an in-depth explanation of AI App Factory's core mechanisms and advanced features, including the detailed working principles of the 7-stage pipeline, Sisyphus scheduler scheduling strategies, permission and security mechanisms, failure handling strategies, and how to optimize context to save token costs.

Prerequisites

Before studying this chapter, please ensure you have completed:

• Quick Start and Installation & Configuration
• 7-Stage Pipeline Overview
• Platform Integration configuration

Chapter Contents

This chapter covers the following topics:

7-Stage Pipeline Deep Dive

• Stage 1: Bootstrap - Structure Product Idea

  • Learn how to transform vague product ideas into structured documentation
  • Understand the input and output formats of the Bootstrap Agent

• Stage 2: PRD - Generate Product Requirements Document

  • Generate MVP-level PRD, including user stories, feature lists, and non-goals
  • Master requirement decomposition and prioritization techniques

• Stage 3: UI - Design Interface and Prototype

  • Use ui-ux-pro-max skill to design UI structure and previewable prototypes
  • Understand interface design workflow and best practices

• Stage 4: Tech - Design Technical Architecture

  • Design minimal viable technical architecture and Prisma data models
  • Master technology selection and architecture design principles

• Stage 5: Code - Generate Runnable Code

  • Generate frontend and backend code, tests, and configurations based on UI Schema and Tech design
  • Understand code generation workflow and template system

• Stage 6: Validation - Validate Code Quality

  • Validate dependency installation, type checking, Prisma schema, and code quality
  • Master automated quality checking workflow

• Stage 7: Preview - Generate Deployment Guide

  • Generate complete running instructions and deployment configurations
  • Learn CI/CD integration and Git Hooks configuration

Internal Mechanisms

• Sisyphus Scheduler Deep Dive

  • Understand how the scheduler coordinates the pipeline, manages state, and executes permission checks
  • Master scheduling strategies and state machine principles

• Context Optimization: Per-Session Execution

  • Learn how to use the factory continue command to save tokens
  • Master best practices for creating a new session at each stage

• Permissions and Security Mechanisms

  • Understand capability boundary matrix, privilege escalation handling, and security check mechanisms
  • Master security configuration and permission management

• Failure Handling and Rollback

  • Learn failure detection, retry mechanisms, rollback strategies, and manual intervention workflows
  • Master troubleshooting and recovery techniques

Learning Path Recommendations

Recommended Learning Order

1. Complete the 7-Stage Pipeline First (in order)

   • Bootstrap → PRD → UI → Tech → Code → Validation → Preview
   • Each stage has clear inputs and outputs; learning in order builds a complete understanding

2. Then Learn the Scheduler and Context Optimization

   • Understand how Sisyphus coordinates these 7 stages
   • Learn how to optimize context to save token costs

3. Finally Learn Security and Failure Handling

   • Master permission boundaries and security mechanisms
   • Understand failure scenarios and response strategies

Focus Areas for Different Roles

Role	Focus Chapters
Developer	Code, Validation, Tech, Orchestrator
Product Manager	Bootstrap, PRD, UI, Preview
Tech Lead	Tech, Code, Security, Failure Handling
DevOps Engineer	Validation, Preview, Context Optimization

Next Steps

After completing this chapter, you can continue learning:

• FAQ & Troubleshooting - Resolve issues in actual use
• Best Practices - Master tips for efficient Factory usage
• CLI Command Reference - View complete command list
• Code Standards - Understand standards for generated code

---

💡 Tip: If you encounter issues during use, please first check the FAQ & Troubleshooting chapter.