Next-Generation AI Workflow Platform - Build AI pipelines with drag-and-drop and interact in real-time
XGEN is an innovative platform that allows you to build AI applications with a visual workflow editor. You can design LangChain-based AI pipelines simply through drag-and-drop without complex coding, and communicate naturally with AI through real-time chat.
- LangChain Node Support: Rich AI nodes including ChatOpenAI, ChatAnthropic, VectorStore
- Drag & Drop: Intuitive interface for workflow composition
- Real-time Connections: Visual representation of data flow between nodes
- Auto Save: Preserve work content with LocalStorage-based system
- Template System: Pre-built workflow templates provided
- Workflow Integration: Real-time conversation with built workflows
- Typing Animation: Elegant UI for displaying AI responses
- Multi-Workflow: Select from multiple workflows for conversation
- Chat History: Save and manage conversation records
- React Hot Toast: Enhanced user experience with notification system
- Execution Monitoring: Track workflow execution status and performance
- Debug Tools: Detailed logging system for development environment
- Settings Management: Global settings and API key management
- Workflow Playground: Test environment provided
- Completed Workflows: History management of executed tasks
- FastAPI Backend: High-performance API server based on Python
- Turbopack: Fast bundler utilizing Next.js 15
- Modular Design: Scalable component-based structure
- Type Safety: Enhanced development productivity and stability with TypeScript
plateerag/
βββ src/
β βββ app/
β β βββ page.tsx # Main landing page
β β βββ layout.tsx # Root layout
β β βββ globals.css # Global styles
β β βββ HomePage.module.scss # Homepage-specific styles
β β βββ config.js # Configuration file (API_BASE_URL, etc.)
β β βββ canvas/ # π¨ Visual workflow editor
β β β βββ page.tsx # Canvas main page
β β β βββ types.ts # Canvas type definitions
β β β βββ components/ # Canvas core components
β β β β βββ Canvas.tsx # Main canvas (ReactFlow-based)
β β β β βββ Node.tsx # AI node component
β β β β βββ Edge.tsx # Connection line component
β β β β βββ Header.tsx # Canvas header (save/load)
β β β β βββ SideMenu.tsx # Side menu container
β β β β βββ ExecutionPanel.tsx # Workflow execution panel
β β β βββ constants/ # Node definitions and constants
β β β β βββ nodes.js # LangChain node data
β β β β βββ workflow/ # Workflow templates
β β β βββ assets/ # Canvas styles (SCSS Modules)
β β βββ chat/ # π¬ AI chat interface
β β β βββ page.tsx # Chat main page
β β β βββ components/ # Chat-related components
β β β β βββ ChatInterface.tsx # Main chat UI
β β β β βββ ChatContent.tsx # Chat content display
β β β β βββ WorkflowSelection.tsx # Workflow selector
β β β βββ assets/ # Chat styles
β β βββ main/ # π Integrated management center
β β β βββ page.tsx # Management center main
β β β βββ components/ # Management tool components
β β β β βββ MainPageContent.tsx # Main dashboard
β β β β βββ Sidebar.tsx # Sidebar navigation
β β β β βββ ContentArea.tsx # Content area
β β β β βββ Executor.tsx # Workflow executor
β β β β βββ Monitor.tsx # Execution monitoring
β β β β βββ Settings.tsx # Settings management
β β β β βββ Playground.tsx # Test environment
β β β β βββ ConfigViewer.tsx # Configuration viewer
β β β β βββ CompletedWorkflows.tsx # Completed workflows
β β β β βββ CanvasIntroduction.tsx # Canvas introduction
β β β βββ assets/ # Management center styles
β β βββ api/ # π API client
β β β βββ workflowAPI.js # Workflow execution API
β β β βββ chatAPI.js # Chat API
β β β βββ nodeAPI.js # Node management API
β β β βββ configAPI.js # Configuration API
β β βββ data/ # π Data management
β β β βββ chatData.js # Chat data model
β β βββ utils/ # π οΈ Common utilities
β β β βββ logger.ts # Debug logger system
β β β βββ generateSha1Hash.ts # Hash generator
β β β βββ debug-guide.js # Debug guide
β β βββ _common/ # Common components
β β βββ components/
β β βββ ToastProvider.jsx # Notification system
β β βββ nodeHook.ts # Node management hook
β β βββ sidebarConfig.ts # Sidebar configuration
β β βββ workflowStorage.js # Workflow storage
β βββ public/ # Static files (icons, images)
βββ package.json # Project configuration and dependencies
βββ next.config.ts # Next.js configuration
βββ tsconfig.json # TypeScript configuration
βββ eslint.config.mjs # ESLint configuration
βββ postcss.config.mjs # PostCSS configuration
βββ DEBUG_GUIDE.md # Debug system usage guide
βββ README.md # Project documentation
XGEN supports various AI nodes from the LangChain ecosystem:
- ChatOpenAI: Support for GPT-4o, GPT-4, GPT-3.5 Turbo
- ChatAnthropic: Claude model series
- Temperature Control: Control creativity and consistency
- Stop Sequence: Output control options
- LLMChain: Basic language model chain
- ConversationChain: Conversational chain
- Agent: Autonomous AI agent
- Tools: External tool integration
- VectorStore: Vector-based document storage
- Memory: Conversation memory management
- Document Loaders: Load various document formats
- Input/Output: Data input/output nodes
- Transform: Data transformation nodes
- Conditional: Conditional branching nodes
- Node.js 18.17 or higher
- npm, yarn, pnpm or bun
# Clone repository
git clone https://github.com/your-org/plateerag.git
cd plateerag
# Install dependencies
npm install
# or
yarn install
# Run development server
npm run dev
# or
yarn devMeet XGEN at http://localhost:3000! π
- Access Canvas Editor (
/canvas) - Select AI Node from left node panel
- Place on canvas with drag & drop
- Create connections between nodes to compose workflow
- Save and run test
- Access Chat Interface (
/chat) - Click Select Workflow button
- Choose desired workflow
- Start natural language conversation
- Check AI responses in real-time
- Access Management Center (
/main) - Check execution status in Performance Dashboard
- Analyze debugging information in Execution Logs
- Optimize workflows in Settings
# Development server (using Turbopack)
npm run dev
# Production build
npm run build
# Run production server
npm run start
# Code linting
npm run lint
# Auto fix code
npm run lint:fix
# Code formatting
npm run format- Next.js 15.3.2: Full-stack React framework based on App Router
- React 19: Utilizing latest React features (Concurrent Features, Server Components)
- TypeScript: Enhanced type safety and development productivity
- SCSS Modules: Component-based style encapsulation
- React Icons: Utilizing Feather Icons set
- React Hot Toast: Elegant notification system
- Turbopack: Fast bundler for Next.js 15 (dev mode)
- ESLint: Code quality and consistency management
- Prettier: Automatic code formatting
- Husky: Quality management through Git hooks
- FastAPI: High-performance API server based on Python
- LangChain: Framework for AI chain composition
- RESTful API: Standard HTTP API communication
- LocalStorage: Client-side data storage
- React State: Application state management
- JSON: Data serialization and API communication
- Use TypeScript strict mode
- ESLint + Prettier automatic formatting
- Style encapsulation with SCSS Modules
- Component-based architecture
XGEN provides a smart debug logger:
import { devLog, prodLog } from '@/app/utils/logger';
// Output only in development environment
devLog.log('Debug information');
devLog.error('Development error');
// Always output (for critical errors)
prodLog.error('Critical error');// Force enable debug logs
enableDebugLogs();
// Disable debug logs
disableDebugLogs();
// Reset to environment settings
resetDebugLogs();
// Check current environment info
checkEnvironment();For detailed usage, refer to DEBUG_GUIDE.md.
| Page | Path | Description |
|---|---|---|
| Home | / |
Project introduction and main landing |
| Canvas | /canvas |
Visual workflow editor |
| Chat | /chat |
AI workflow chat interface |
| Management | /main |
Workflow management and monitoring |
Thank you for contributing to the XGEN project!
- Fork this repository
- Create Feature branch (
git checkout -b feature/amazing-feature) - Commit changes (
git commit -m 'Add amazing feature') - Push to branch (
git push origin feature/amazing-feature) - Create Pull Request
# Install development dependencies
npm install
# Setup pre-commit hooks
npm run prepare
# Check code quality
npm run lint
npm run formatThis project follows GPL-3.0.
- Plateer AI-LAB
- CocoRoF - Jang Ha-ryeom
- haesookimDev - Kim Hae-su
- SonAIengine - Son Seong-jun (AI-LAB Part Leader)
- daunle - Lee Da-un
- GitHub Repository: XGEN Repository
- Development Documentation:
DEBUG_GUIDE.md- Debug system usage guide - Issue Reports: Bug reports and feature requests through GitHub Issues
- Tech Blog: Plateer Tech Blog
Made with β€οΈ by Plateer AI-LAB
β Star this repo β’ π Report Bug β’ π‘ Request Feature






