A full-featured Agent chatbot built with Next.js 15 and iFlow CLI SDK, providing Web-based iFlow-like interactive experience.
π δΈζζζ‘£ (Chinese Documentation)
Great thanks to the iFlow team for their excellent work and all staff for their hard work! πππ
iFlow Chatbot is a production-ready Agent chat application that integrates the iFlow CLI SDK to provide powerful features:
- π€ Multi-model AI support (MiniMax-M2, Qwen3, DeepSeek, GLM, Kimi)
- π¬ Intelligent conversation with context memory
- π§ Tool call visualization and task planning
- π± Fully responsive design (Desktop + Mobile)
- π¨ Dark mode with system theme detection
- π Secure authentication system
- πΎ Persistent chat history and workspace management
- β‘ High-performance React rendering optimization
- Multi-turn dialogue with context understanding
- Real-time streaming message transmission
- Markdown rendering with code highlighting
- Tool call status visualization
- Task plan progress tracking
- File operations and code generation
- Visual tool call process display
- Task execution plan presentation
- Code editor and file preview
- Fully responsive design
- Desktop: Sidebar + main chat area
- Mobile: Drawer-style sidebar + fullscreen chat
- Dark mode support
- Touch-optimized interface
Supported Models:
- MiniMax-M2
- Qwen3-Coder-Plus
- DeepSeek-V3.2
- GLM-4.6
- Kimi-K2-0905
Permission Modes:
default: Confirm every operationautoEdit: Auto-execute edit operationsyolo: Auto-execute all operationsplan: Plan before execution
- Framework: Next.js 15 (App Router)
- Language: TypeScript
- UI Components: shadcn/ui + Tailwind CSS
- State Management: React Hooks + Context
- Animations: Framer Motion
- Code Highlighting: react-syntax-highlighter
- Database: PostgreSQL + Drizzle ORM
- Authentication: NextAuth v5
- API: Next.js API Routes
- Real-time Communication: Server-Sent Events (SSE)
- AI SDK: iFlow CLI SDK v0.1.2
- Message System: WebSocket + SSE
- Workspace Management: File system integration
- Session Management: 30-minute auto-cleanup
The simplest way: You can directly hand the current project to iFlow for review, and it will automatically analyze the project structure and complete the deployment. No manual environment variable configuration or installation commands required.
If manual installation is needed, ensure the following environment requirements:
- Node.js 18+
- PostgreSQL database
- iFlow API key
Installation Steps:
- Clone the project
git clone https://github.com/stone100010/iflow-chatbot.git
cd iflow-chatbot- Install dependencies
pnpm install- Configure environment variables
Create .env.local file:
# Database
POSTGRES_URL="postgresql://..."
# iFlow API
IFLOW_API_KEY="sk-..."
IFLOW_BASE_URL="https://apis.iflow.cn/v1"
# NextAuth
AUTH_SECRET="your-secret-key"- Database migration
pnpm db:migrate- Start development server
pnpm devThe app will start at localhost:3000.
- Desktop: Sidebar + main chat area layout
- Mobile: Drawer-style sidebar + fullscreen chat experience
- Adaptive: Config selector auto-switches between Popover/Sheet
- next-themes integration
- System theme auto-detection
- Dynamic theme color switching
- Streaming message display
- Tool call progress animations
- Task plan progress bars
- Collapsible think blocks
- Code syntax highlighting (oneDark theme)
- NextAuth v5 integration
- Guest user support (
guest-{timestamp}) - Session management
- API key environment variable protection
- default: Each operation requires confirmation
- autoEdit: Auto-execute edit operations
- yolo: Auto-execute all operations
- plan: Make detailed plans before execution
Contributions are welcome! Please feel free to submit Pull Requests.
- Fork this repository
- Create a feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'feat: add some amazing feature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
This project is licensed under the Apache-2.0 License.
- Vercel for Next.js and deployment platform
- shadcn/ui for beautiful components
- All contributors and supporters
For questions or feedback, please create an Issue on GitHub.
Project Link: https://github.com/stone100010/iflow-chatbot