Hasilnya lebih bagus di local daripada di production
Prompt sama, code sama, hasil generate beda, anda aja bingung, apalagi saya
- AI Chat Generation: Generate kode menggunakan berbagai model LLM (GPT-4)
- Code Sandbox: Jalankan kode secara aman di environment terisolasi menggunakan E2B
- Real-time Preview: Lihat hasil kode langsung di browser
- Rate Limiting: Kontrol penggunaan API dengan sistem rate limiting
- Authentication: Sistem autentikasi menggunakan Clerk
- Framework: Next.js 14 dengan App Router
- AI/LLM: Vercel AI SDK, OpenAI
- Sandbox: E2B Code Interpreter
- Authentication: Clerk
- Styling: Tailwind CSS + shadcn/ui
- Monitoring: LangSmith
- Database: Supabase (opsional)
src/
├── app/
│ ├── api/
│ │ └── ai/
│ │ ├── route.ts # Main router (backward compatibility)
│ │ ├── chat/route.ts # AI chat endpoint
│ │ └── sandbox/route.ts # Code execution endpoint
│ ├── (auth)/ # Authentication pages
│ ├── (landingpage)/ # Landing page
│ └── main/ # Main application
├── components/ # Reusable UI components
├── lib/ # Utilities & configurations
├── inngest/ # Background job functions
└── types/ # TypeScript type definitions
Distributed under the MIT License. See LICENSE
for more information.
Wahyu - @wahyudesu