https://github.com/operatornext/operatornext
OperatorNext is an AI-powered intelligent browser automation platform that understands and executes complex browser tasks through natural language processing.
https://github.com/operatornext/operatornext
ai automation browser-automation fastapi langchain nextjs playwright react typescript websocket
Last synced: 3 months ago
JSON representation
OperatorNext is an AI-powered intelligent browser automation platform that understands and executes complex browser tasks through natural language processing.
- Host: GitHub
- URL: https://github.com/operatornext/operatornext
- Owner: OperatorNext
- License: mit
- Created: 2025-02-03T17:49:41.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2025-03-05T18:31:16.000Z (4 months ago)
- Last Synced: 2025-03-05T19:27:01.364Z (4 months ago)
- Topics: ai, automation, browser-automation, fastapi, langchain, nextjs, playwright, react, typescript, websocket
- Language: TypeScript
- Homepage: https://operatornext.com
- Size: 3.24 MB
- Stars: 9
- Watchers: 1
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Citation: CITATION.cff
- Codeowners: CODEOWNERS
- Security: SECURITY.md
Awesome Lists containing this project
README
# OperatorNext ๐ค
[](https://github.com/OperatorNext/OperatorNext/blob/main/LICENSE)
[](https://github.com/OperatorNext/OperatorNext/stargazers)
[](https://github.com/OperatorNext/OperatorNext/issues)
[](https://github.com/OperatorNext/OperatorNext/pulls)
[](https://discord.gg/zafb9TzYYA)
[](https://github.com/OperatorNext/OperatorNext/releases)
[English](./README.md) | [็ฎไฝไธญๆ](./README.zh-CN.md)
---
**Open-Source Alternative to OpenAI Operator**
*LLM-Powered Browser Automation Agent*
[Get Started](#-getting-started) โข [Documentation](https://github.com/OperatorNext/OperatorNext/tree/main/docs) โข [Examples](#-usage-example) โข [Contributing](#-contributing) โข [Demo](https://operatornext.com)
OperatorNext is an open-source AI agent platform that understands and executes complex browser tasks through natural language processing and visual reasoning. By combining state-of-the-art LLM technologies (including GPT-4o, Claude, and more) with browser automation, we provide developers and users with a powerful Computer-Using Agent (CUA) for web automation, data collection, UI testing, and various other scenarios.
๐ฅ๏ธ Modern Interface with Real-time Task Monitoring
๐ค Intelligent Task Execution with Visual Feedback
> โ ๏ธ **Project Status**
>
> This project is in early development stage. Core features are under active development and not yet implemented.
>
> Please note that breaking changes may occur frequently during this phase.
### ๐ Why Choose OperatorNext?
| Feature | OperatorNext | OpenAI Operator |
|---------|-------------|-----------------|
| License | MIT Open Source | Proprietary |
| Deployment | Self-hosted & Cloud | Cloud-only |
| Data Privacy | Local Processing | Cloud Processing |
| Customization | Full Control | Limited |
| Cost | Free & Self-hosted | Usage-based Pricing |
| API Integration | Flexible & Open | Restricted |
OperatorNext empowers developers with:
- ๐ **Web Scraping & Data Extraction** - Automated data collection with pixel-level accuracy
- ๐งช **End-to-End Testing** - Modern alternative to Selenium for UI/UX testing
- ๐ค **RPA (Robotic Process Automation)** - Chain-of-Thought planning for complex tasks
- ๐ **Web Testing & QA** - Visual reasoning based quality assurance
- ๐ **Data Mining & Analytics** - Intelligent web data gathering with self-correction
- ๐ **Workflow Automation** - Custom workflow design with plugin ecosystem
Perfect for:
- DevOps and QA Teams (Automated Testing)
- Data Scientists and Researchers (Web Scraping)
- Digital Marketing Professionals (Form Automation)
- Business Process Automation (RPA Solutions)
- Web Developers and Testers (GUI Testing)
- Enterprise Automation Solutions (Custom Workflows)
## โจ Features
- ๐ค **AI Agent & Visual Reasoning** - Complete complex browser operations through natural language and visual understanding, powered by GPT-4o multimodal capabilities
- ๐ง **Chain-of-Thought Planning** - Advanced task planning and execution with reinforcement learning for optimal automation
- ๐ฏ **Precise GUI Interaction** - Pixel-perfect DOM operations, XPath navigation, and complex interaction scenarios using computer vision
- ๐ **Real-time Task Tracking** - WebSocket-based monitoring system with CPU, memory, and network metrics for execution insights
- ๐ **Privacy-First Design** - Local processing of sensitive data with comprehensive error handling and self-correction mechanisms
- ๐ **Cross-Platform & Multilingual** - Full i18n support with Chinese/English interfaces, works on Windows, macOS, and Linux
- ๐ **Extensible Architecture** - REST API, WebSocket endpoints, and plugin system for seamless integration
- ๐ **Cloud & Self-Hosted** - Deploy on your infrastructure or use our cloud solution for maximum flexibility
- โก **High Performance** - Parallel task execution with optimized resource management
- ๐จ **Modern Developer Experience** - Beautiful UI/UX built with Next.js and Tailwind CSS, extensive API documentation
## ๐ Getting Started
Try our online demo at [operatornext.com](https://operatornext.com) or set up your own instance:
### Prerequisites
- Docker & Docker Compose
- Node.js 18+
- pnpm 10+
- Chrome/Chromium browser
### Installation
1. Clone the repository
```bash
git clone https://github.com/OperatorNext/OperatorNext.git
cd OperatorNext
```
2. Copy environment variable templates
```bash
# Copy frontend environment variables
cp frontend/.env.local.example frontend/.env.local
# Copy Docker environment variables
cp .env.example .env
```
3. Install frontend dependencies
```bash
cd frontend
pnpm install
```
4. Initialize database and generate types
```bash
# Push database schema
sudo pnpm db:push
# Generate Prisma client and types
sudo pnpm db:generate
```
> Note: `sudo` might be required for database operations depending on your system configuration.
### Start Services
1. Start Docker services
```bash
docker-compose up -d
```
This will start the following services:
| Service | URL | Description |
|---------|-----|-------------|
| Web Application | http://localhost:3000 | Next.js frontend application |
| PgAdmin | http://localhost:5051 | PostgreSQL database management |
| Maildev | http://localhost:8026 | Email testing interface |
| MinIO Console | http://localhost:9003 | Object storage management |
| MinIO API | http://localhost:9002 | S3-compatible API endpoint |
| PostgreSQL | localhost:5438 | Database (connect via psql or GUI) |
### Default Credentials
> โ ๏ธ These are development credentials. Do NOT use in production!
- **PostgreSQL**:
- User: operatornext_prod_user
- Database: operatornext_production
- **PgAdmin**:
- Email: [email protected]
- Password: See `.env` file
- **MinIO**:
- Access Key: See `MINIO_ROOT_USER` in `.env`
- Secret Key: See `MINIO_ROOT_PASSWORD` in `.env`
2. Start frontend development server
```bash
cd frontend
pnpm dev
```
Visit http://localhost:3000 to use the application.
## ๐ Usage Example
```python
# Create a new browser task
task = {
"task_description": "Login to GitHub and star a repository"
}
response = requests.post("http://localhost:8000/api/tasks", json=task)
task_id = response.json()["task_id"]
# Monitor task status via WebSocket
ws = websockets.connect(f"ws://localhost:8000/ws/tasks/{task_id}")
```
For more examples, please visit our [documentation](https://github.com/OperatorNext/OperatorNext/tree/main/docs).
## ๐ง Technology Stack
### AI & Automation
- LLM Support - Compatible with GPT-4o, Claude, and other language models
- LangChain - Large Language Model (LLM) orchestration framework
- Computer Vision - Pixel-level DOM interaction and visual analysis
- Reinforcement Learning - Self-improving task execution strategies
- Chain-of-Thought - Advanced planning and decision making
### Backend Infrastructure
- FastAPI - High-performance Python web framework for building scalable APIs
- WebSocket - Real-time bidirectional communication for task monitoring
- Playwright - Modern web testing and automation with superior stability
- PostgreSQL - Advanced open-source database for task management
- MinIO - S3-compatible object storage for artifact management
- Redis - In-memory data structure store for caching and queuing
- Docker - Containerization and deployment automation
### Frontend Technologies
- Next.js 15 (App Router) - React framework with server-side rendering
- React 19 - Latest version with concurrent features and Suspense
- TypeScript - Type-safe JavaScript development for reliability
- Tailwind CSS - Utility-first CSS framework for modern UI
- Shadcn UI - Modern and accessible component library
- Prisma - Next-generation ORM for type-safe database access
- Turbo Repo - High-performance monorepo build system
- WebSocket - Real-time updates and task monitoring
- Biome - Fast and reliable code formatter
### DevOps & Quality
- Docker Compose - Multi-container orchestration
- GitHub Actions - CI/CD automation pipeline
- Playwright - End-to-end testing framework
- Prisma - Database schema management and migrations
- Biome - Code quality and formatting tools
- pnpm - Fast, disk space efficient package manager
### Security & Privacy
- Local Processing - Sensitive data handling
- End-to-End Encryption - Secure communication
- Role-Based Access - Fine-grained permissions
- Audit Logging - Comprehensive activity tracking
## ๐ Documentation
For detailed documentation, please visit our [documentation](https://github.com/OperatorNext/OperatorNext/tree/main/docs).
## ๐ค Contributing
We welcome all forms of contributions, whether it's new features, documentation improvements, or bug reports. Please check our [Contributing Guide](CONTRIBUTING.md) for more information.
## ๐ License
This project is licensed under the [MIT](LICENSE) License.
## ๐ Acknowledgments
This project is inspired by and built upon:
- [browser-use](https://github.com/browser-use/browser-use)
- [browserless](https://github.com/browserless/browserless)
Thanks to all the developers who have contributed to this project!
## ๐ Star History
## ๐ฎ Contact & Community
Join our growing community:
- [Discord](https://discord.gg/zafb9TzYYA) - Join our community for discussions, support, and updates
- [Slack](https://join.slack.com/t/operatornext/shared_invite/zt-2yzynnxiv-ywt7Z8UtykGAm6EUfpljQA) - Join our Slack workspace for team collaboration
- [GitHub Issues](https://github.com/OperatorNext/OperatorNext/issues) - Bug reports and feature requests
- [GitHub Discussions](https://github.com/OperatorNext/OperatorNext/discussions) - Technical discussions and questions
- Email: [email protected]
- Telegram: [@HaiPro_2025](https://t.me/HaiPro_2025)
Company: CyberPoet LLC (Position: CEO)