GreenBuild AI

GreenBuild AI is a full-stack sustainable construction workflow app built with React 18, Vite, Tailwind CSS, FastAPI, Gemini 1.5 Pro, Open-Meteo, fpdf2, and local JSON storage.

Features

• 5-step project intake form for building parameters
• Async backend analysis jobs with polling
• Open-Meteo climate enrichment
• Gemini 1.5 Pro structured material recommendations across 10 building components
• Cinematic dark dashboard with charts and component drill-down
• PDF report download
• Floating follow-up chat widget with streaming responses

Project Structure

greenbuild-ai/
  backend/
  frontend/
  .env
  .vscode/tasks.json

Setup

1. Backend

Navigate to the backend directory:

cd backend

Create and activate a virtual environment:

python -m venv .venv
.\.venv\Scripts\activate

Install dependencies:

pip install -r requirements.txt

Set your Gemini key in the .env file at the root of the project.

Start the API:

python -m uvicorn app.main:app --reload --host 127.0.0.1 --port 8001

2. Frontend

Navigate to the frontend directory:

cd frontend

Install dependencies:

npm install

Start the development server:

npm run dev

The Vite app runs on http://localhost:5173 and expects the API at http://localhost:8000.

VS Code

Press Ctrl+Shift+B and choose the default build task to launch both servers in parallel using .vscode/tasks.json.

API Routes

• POST /analyze
• GET /status/{job_id}
• GET /results/{slug}
• GET /report/{slug}
• POST /chat/{slug}
• GET /climate?location=...

Notes

• If GEMINI_API_KEY is missing, the backend falls back to a deterministic mock analysis so the app still runs locally.
• Analysis and chat results are persisted in db.json after first run.