Current build

📝: (Select an Ollama model before prompting the agent!)

• Next.js app router with React/Tailwind
• Some components from Llamacoder
• Code sandbox by Sandpack
• Electron builder by OpenJS Foundation.
• Token stats & app settings by nexaforge-dev

Features

📝: (System prompt is currently preset for creating a React-based App.tsx codebase)

• Code + Preview frames
• Chat, Settings & Token frames
• Builds as an Electron app, or runs in a browser

Local LLMs

📝: (qwen2.5-coder:7b is currently hardcoded in actions.tsx)

• Ollama server (required)
• Qwen models by Alibaba (optional)
• Deepseek R1 (pending)

Prep / Planning

• 🔌 dark mode
• 🔌 draggable frames
• 🔌 remember chosen model
• 🔌 add resize/close buttons for Electron app
• 🔌 thinking frame for Qwen Coder / Deepseek R1

Cloning & Building

1. Clone repo: git clone https://github.com/kontains/volt
2. Start Ollama server on your machine.
3. Run npm install --legacy-peer-deps

run in a browser: npm run dev then open a browser at: http://localhost:3000/?t=1
📝: (Defaults to first available port after 3000 if busy)

or build and run the app: npm run build & npm run electron-dev 📝: (Production build is a WIP)

Contributing

• Issues and PRs are open
• 

---

(sample)

---

---