Skip to content

Repository files navigation

Domain Flow

Software Development Design Flow methodology for AI-assisted development.

Overview

Domain-Driven Development (DDD) methodology with hybrid Waterfall-Agile approach. Build applications with AI coding agents through structured documentation and domain-driven design.

Quick Start

# Set up prompts
export SYMLINKS_TARGET_DIR="~/your-project/.claude/commands"
ln -s ./prompts/* $SYMLINKS_TARGET_DIR/

# Set up templates
export SYMLINKS_TARGET_DIR="~/your-project/docs"
ln -s ./templates/* $SYMLINKS_TARGET_DIR/

# Set up agents
export SYMLINKS_TARGET_DIR="~/your-project/.claude/agents"
ln -s ./agents/* $SYMLINKS_TARGET_DIR/

Core Components

7-Step Workflow

graph TD
    S1["Step 1: Discovery & Concept"] -->
    S2["Step 2: High-Level System Design"] -->
    S3["Step 3: Feature Roadmap"] -->
    S4["Step 4: Foundation & Setup"] -->
    S5["Step 5: Design"] -->
    S6["Step 6: Implementation"] -->
    S7["Step 7: Phase Completion"]
Loading

Key Features

Domain-Driven Design

  • Dynamic domain discovery via Context.toml
  • Domain isolation with clear boundaries
  • GRAPH_TAG system for component tracking
  • 4 DTO pattern for frontend/backend communication

AI Agent Integration

  • Context analysis and architecture discovery
  • Specialized prompts for each development phase
  • Dynamic workspace configuration
  • Template-driven documentation generation

Usage

  1. Discover domains dynamically
  2. Use prompts for each development phase
  3. Follow templates for consistent documentation
  4. Leverage agents for architecture analysis

Project Structure

domain-flow/
├── workflow.md                  # Main methodology
├── prompts/                     # AI prompts (17 files)
├── templates/                   # Document templates (22 files)
├── agents/                      # AI agents
├── STEERING.md                  # AI context instructions
└── docs/domains/                # Domain documentation

Getting Started

  1. Clone the repository
  2. Set up symbolic links using commands above
  3. Follow Step 1.1 for new projects

License

MIT License

About

AI-assisted development methodology combining Domain-Driven Design with structured 7-step workflow. Enables AI coding agents to build complex applications through domain isolation, specialized prompts, and template-driven documentation.

Resources

Stars

4 stars

Watchers

0 watching

Forks

Contributors

Languages