Skip to content

Repository files navigation

Vibe Cover

Vibe

A lightweight clone of Lovable built to explore how AI-powered application generation works under the hood.

Vibe is an experimental platform that allows users to generate, edit, and execute code inside a controlled sandbox environment using natural language prompts.


Overview

The goal of this project is not to replicate Lovable entirely, but to understand and rebuild its core mechanics:

  • Prompt-driven code generation
  • Secure code execution in an isolated environment
  • Real-time interaction between user input and generated output
  • Scalable fullstack architecture

This project is part of a broader exploration into AI-assisted development tools.


Tech Stack

Frontend

  • Next.js 15 (App Router, Turbopack)
  • React 19
  • Tailwind CSS v4
  • shadcn/ui and Radix UI
  • React Hook Form with Zod validation

Backend

  • tRPC for end-to-end type-safe APIs
  • Prisma ORM

Infrastructure & Workflows

  • Inngest for background jobs and event-driven workflows

Authentication

  • Clerk

AI & Code Execution

  • E2B Code Interpreter (Dockerized sandbox environment)
  • Inngest Agent Kit

Data Management

  • TanStack React Query
  • SuperJSON

Features

  • Prompt to code generation
  • Interactive code editor running in a sandbox
  • Secure execution environment using Docker
  • Authentication and session management
  • Background processing for AI workflows
  • Modular and maintainable architecture

Architecture

The system is built around a modern fullstack approach:

  • The frontend communicates with the backend using tRPC
  • AI-related tasks are processed asynchronously via Inngest
  • Code execution is isolated using a Dockerized E2B sandbox
  • Prisma handles database interactions
  • Clerk manages authentication and user sessions

Getting Started

git clone https://github.com/Edwinfom00/vibe.git
cd vibe

npm install

cp .env.example .env

npm run dev

About

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages