Legacy.Ai

Legacy.Ai is a small example project that demonstrates using a backend Express API to call a Gemini generative model for automated code review, with a lightweight React + Vite frontend to submit code and display the review.

This repository has two main pieces:

• BackEnd/: Node.js + Express server that wraps the Gemini generative model and exposes an API endpoint.
• FrontEnd/: React (Vite) app that provides a code editor and displays the model's review.

Quick architecture

• FrontEnd (Vite + React) sends POST requests to the BackEnd endpoint /ai/get-review with a JSON body { code: string }.
• BackEnd forwards the prompt to the Gemini generative model via @google/generative-ai and returns the model's text response to the FrontEnd.

What it's useful for

• Prototyping automated code review features.

Running locally

Prerequisites:

• Node.js (16+ recommended)
• npm
• A Gemini API key (set as GEMINI_API_KEY in the environment)

1. BackEnd

cd BackEnd
npm install
# Set the GEMINI_API_KEY for the current PowerShell session
$env:GEMINI_API_KEY = "your_gemini_api_key_here"
# Start the backend (server listens on port 3000)
node server.js

Notes:

• The Gemini API key is read from process.env.GEMINI_API_KEY.

2. FrontEnd

cd FrontEnd
npm install
npm run dev

• The FrontEnd uses Vite. By default it will run on http://localhost:5173 (or another free port).
• The app posts to http://localhost:3000/ai/get-review in development. If you change backend port, update the frontend request URL in FrontEnd/src/App.jsx.

Environment variables

• GEMINI_API_KEY — API key for Google Gemini/Generative AI. Required by the backend.

Troubleshooting

• If the frontend cannot reach the backend, ensure the backend is running on port 3000 and that CORS is enabled (the backend already uses cors() in src/app.js).
• If the model call fails, check that GEMINI_API_KEY is set and valid.

Connect with me

• LinkedIn: https://www.linkedin.com/in/devadi
• GitHub: https://github.com/ADI-2707