AI-powered platform that analyzes architecture, development workflows, and engineering documentation.
GoodOne demonstrates what happens when AI becomes part of the engineering workflow itself.
Instead of only generating code, the platform analyzes:
• architecture documentation
• development workflows
• project knowledge
• engineering risks
Live demo
https://goodone.ch
This project is a showcase of AI-driven software development.
Highlights:
• 100% of the source code was generated by AI
• Iteration planning assisted by AI
• Architecture design refined with AI
• AI features operate directly at runtime
GoodOne is not yet a finished product but a working demonstration of AI-assisted software engineering.
The platform is under active development, and new AI capabilities are continuously added.
| Feature | Description |
|---|---|
| Architecture Q&A | Ask natural-language questions about system architecture |
| AI Risk Radar | Detect recurring engineering and delivery risks |
| Sprint Retrospective | Generate retrospective insights automatically |
| ADR Drift Detection | Detect architectural drift from ADR decisions |
| AI Task Parsing | Convert natural language into structured tasks |
| AI Economy | Track AI usage, credits, and operational costs |
Angular Frontend
│
▼
Spring Boot API
│
▼
AI Analysis Layer
│
┌─────────┼─────────┐
▼ ▼ ▼
Tasks Documentation ADRs
AI analyzes these artifacts to generate insights about the engineering system.
Ask natural-language questions about the system architecture.
Example questions:
• Which components use AI at runtime?
• How does authentication work?
• How is reCAPTCHA verified?
The AI answers using internal architecture documentation.
Automatically detect recurring engineering risks.
Examples:
• tasks marked DONE but still containing open items
• missing verification sections
• documentation inconsistencies
Helps teams detect systematic quality problems early.
Generate AI-assisted sprint retrospectives based on development tasks.
The system analyzes:
• task completion patterns
• recurring blockers
• documentation quality
Architecture Decision Records define architectural intent.
GoodOne monitors implementation and detects when systems drift away from those decisions.
This helps maintain long-term architectural integrity.
Turn natural language into structured work items.
Example:
Prepare architecture review next Friday
The AI extracts:
• title
• due date
• category
Track and understand AI usage inside the platform.
Examples:
• AI credit requests
• usage dashboards
• cost transparency
Clone the repository
git clone https://github.com/JuergGood/angularai
Start the stack
cp .env.example .env docker compose -f deploy/dev/docker-compose.yml up --build
Application endpoints
Frontend
http://localhost
Backend API
http://localhost:8080/api
Mailpit
http://localhost:8025
Some runtime features require access to an AI model.
Runtime AI features currently use the OpenAI API.
OPENAI_API_KEY=your_key_here
A local runtime using Ollama is currently under development.
This will allow running AI features locally without external APIs.
The application uses PostgreSQL, automatically started via Docker.
Documentation is located in the doc directory.
Key entry points:
Architecture
doc/architecture/index.md
User Guide
doc/user-guide/user-guide.md
Admin Guide
doc/admin-guide/admin-guide.md
Deployment
doc/infrastructure/Deployment.md
GoodOne explores how AI can augment software engineering workflows.
Instead of replacing developers, the platform helps teams:
• understand complex architectures
• detect engineering risks
• analyze development processes
• preserve architectural intent
⭐ If you find this project interesting, please consider starring the repository.