Code Vision App - Work in Progress

A Python-powered AI agent system for querying building codes and regulations with clause-specific expertise.

Architecture

This application consists of:

• FastAPI + LangGraph Backend: Orchestration agent with specialist clause agents (B-H)
• Next.js Frontend: Static chat UI with Supabase Auth
• Supabase pgvector: Vector database for clause embeddings
• Fly.io Deployment: Auto-scaling Python API service
• Netlify Deployment: Global CDN for frontend

Quick Start

Prerequisites

• Python 3.12+
• Node.js 18+
• Poetry
• Docker (optional)

Installation

# Install Python dependencies
make install

# Install frontend dependencies
cd frontend && npm install

Development

# Run backend in development mode
make dev

# Run frontend in development mode (separate terminal)
make frontend-dev

Testing

# Run all tests with coverage
make test

# Run linting
make lint

# Format code
make format

Production Build

# Build Docker image
make docker-build

# Run production container
make docker-run

Project Structure

├── .github/workflows/     # CI/CD pipeline
├── src/agent_project/     # Python backend
│   ├── application/       # FastAPI routers & DI
│   ├── core/
│   │   ├── agents/        # LangGraph agents (orchestrator + specialists)
│   │   ├── prompts/       # Jinja prompt templates
│   │   ├── tools/         # Shared retrieval & reasoning
│   │   └── utils/         # Utilities (logging, timing)
│   ├── infrastructure/
│   │   ├── vector_db/     # Supabase pgvector client
│   │   ├── llm/           # LLM provider wrappers
│   │   └── auth/          # Supabase JWT validation
│   └── config.py          # Pydantic settings
├── frontend/              # Next.js static export
├── tools/                 # CLI utilities
├── tests/                 # Pytest test suites
└── claude-rules/          # Claude AI assistant rules

Environment Variables

Create a .env file with:

# Supabase
SUPABASE_URL=your_supabase_url
SUPABASE_ANON_KEY=your_supabase_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_service_role_key

# LLM Providers
OPENAI_API_KEY=your_openai_key
ANTHROPIC_API_KEY=your_anthropic_key

# Application
APP_ENV=development
LOG_LEVEL=INFO
API_VERSION=v1

Deployment

The application auto-deploys on push to main:

1. Backend: Builds Docker image → pushes to GHCR → deploys to Fly.io
2. Frontend: Builds static export → deploys to Netlify
3. Infrastructure: Terraform Cloud manages secrets and scaling

API Documentation

Once running, visit:

• API Docs: http://localhost:8000/docs
• Health Check: http://localhost:8000/health
• Metrics: http://localhost:8000/metrics

Contributing

1. Create a feature branch
2. Make your changes
3. Run tests and linting: make test lint
4. Submit a pull request