The most advanced open-source platform for creating multi-agent AI workflows with drag-and-drop simplicity.
π― Features β’ π Architecture β’ π Quick Start β’ π Documentation β’ π€ Contributing
π¨ Visual Agent Builder Interface - Create sophisticated AI workflows with drag-and-drop simplicity. The screenshot shows the intuitive Block Palette with 50+ pre-built components, real-time workflow canvas, and seamless agent orchestration capabilities.
Agent Builder is a revolutionary visual AI workflow builder that empowers developers and businesses to create sophisticated multi-agent AI systems without writing complex code. Think of it as the "Zapier for AI Agents" - but with enterprise-grade capabilities, advanced orchestration patterns, and unlimited customization.
The interface above demonstrates Agent Builder's powerful visual workflow builder in action:
- π Left Panel: Comprehensive Block Palette with categorized components
- Agents: General Agent, Control Agent, AI Agent blocks
- Popular Blocks: Most-used components with popularity indicators
- Categories: Organized seor Plugins, Workflows, Triggers, Data & Knowledge
- π¨ Center Canvas: Visual workflow designer showing a sample automation
- Connected Blocks: Email β Database β End workflow with visual connections
- Real-time Status: Live execution indicators and data flow visualization
- π§ Right Panel: Properties and configuration for selected blocks
- π Multi-language: Korean interface demonstrating global accessibility
- π¨ Visual First: Drag-and-drop interface powered by ReactFlow
- π€ Multi-Agent Orchestration: 17 advanced orchestration patterns including 2025 trends
- β‘ Real-time Execution: Live monitoring with Server-Sent Events
- π 50+ Integrations: Connect to any API, database, or service
- π§ Multi-LLM Support: OpenAI, Claude, Gemini, Grok, Ollama, and more
- π Enterprise RAG: Vector search, hybrid search, knowledge graphs
- π Production Ready: Security, monitoring, scaling, and deployment
| Feature | Description |
|---|---|
| Drag & Drop Editor | Intuitive ReactFlow-based visual editor with 50+ pre-built blocks |
| Real-time Preview | See your workflow execute in real-time with live data flow |
| Smart Validation | Intelligent validation with helpful error messages and suggestions |
| Version Control | Built-in versioning with diff visualization and rollback capabilities |
The screenshot above showcases our powerful Agent Builder interface with:
| Component | Description |
|---|---|
| π Block Palette | Comprehensive library of 50+ pre-built blocks organized by category |
| π Block Categories | Agents, Plugins, Workflows, Triggers, Data & Knowledge, Orchestration, and Developers sections |
| π§ Popular Blocks | Quick access to frequently used components like General Agent, Control Agent, and specialized agents |
| π¨ Visual Canvas | Drag-and-drop workflow designer with real-time connection visualization |
| β‘ Live Execution | Real-time workflow execution with visual feedback and status indicators |
| π Smart Connections | Intelligent block linking with automatic data flow validation |
Key Interface Features:
- Korean/English Support: Fully localized interface for global accessibility
- Dark/Light Theme: Professional UI that adapts to your preference
- Responsive Design: Works seamlessly across desktop, tablet, and mobile devices
- Contextual Help: Built-in tooltips and guidance for every component
- Keyboard Shortcuts: Power-user shortcuts for rapid workflow development
| Pattern Category | Patterns | Use Cases |
|---|---|---|
| Core Patterns | Sequential, Parallel, Hierarchical, Adaptive | Basic workflow automation |
| 2025 Trends | Consensus Building, Dynamic Routing, Swarm Intelligence, Event-Driven, Reflection | Advanced AI coordination |
| 2026 Next-Gen | Neuromorphic, Quantum Enhanced, Bio-Inspired, Self-Evolving, Federated, Emotional AI | Cutting-edge AI research |
- Communication: Slack, Discord, Email, SMS, Teams
- Databases: PostgreSQL, MySQL, MongoDB, Redis, Milvus
- APIs: REST, GraphQL, WebSocket, gRPC
- Cloud Services: AWS, GCP, Azure, Cloudflare
- AI Services: OpenAI, Anthropic, Google AI, Cohere, Hugging Face
- Document Processing: PDF, DOCX, PPTX, Excel, OCR
- Web: Web scraping, search engines, social media APIs
- Vector Search: Milvus-powered semantic search with 1000+ embedding models
- Hybrid Search: Combine semantic and keyword search with BM25
- Knowledge Graphs: NetworkX-powered entity relationship mapping
- Document Processing: Advanced PDF, Office, and image processing with OCR
- Multi-modal AI: Support for text, images, audio, and video processing
- Real-time Streaming: Server-Sent Events for live execution monitoring
- Auto-scaling: Dynamic resource allocation based on workload
- Caching: Redis-powered intelligent caching for optimal performance
- Load Balancing: Built-in load balancing for high-availability deployments
- Monitoring: Comprehensive metrics, logging, and error tracking
graph TB
subgraph "Frontend Layer"
A[Next.js 15 + React 19]
B[ReactFlow Visual Editor]
C[Real-time Dashboard]
D[Plugin Management]
end
subgraph "API Gateway"
E[FastAPI + WebSocket]
F[Authentication & Authorization]
G[Rate Limiting & Security]
end
subgraph "Core Engine"
H[Workflow Execution Engine]
I[Multi-Agent Orchestrator]
J[Plugin System]
K[Event Bus]
end
subgraph "AI & ML Layer"
L[LangChain + LangGraph]
M[Multi-LLM Router]
N[Vector Embeddings]
O[Knowledge Graph]
end
subgraph "Data Layer"
P[PostgreSQL]
Q[Milvus Vector DB]
R[Redis Cache]
S[File Storage]
end
A --> E
B --> E
C --> E
D --> E
E --> H
F --> H
G --> H
H --> I
I --> J
J --> K
H --> L
L --> M
M --> N
N --> O
H --> P
H --> Q
H --> R
H --> S
- Framework: Next.js 15 with React 19 and TypeScript
- UI Components: Radix UI + Tailwind CSS + Framer Motion
- Visualization: ReactFlow for workflow diagrams, D3.js for analytics
- State Management: Zustand + TanStack Query for server state
- Real-time: Server-Sent Events + WebSocket connections
- Framework: FastAPI with async/await support
- AI/ML: LangChain, LangGraph, LiteLLM for multi-provider support
- Vector Search: Milvus with sentence-transformers embeddings
- Database: PostgreSQL with SQLAlchemy 2.0 and Alembic migrations
- Caching: Redis for session management and result caching
- Security: JWT authentication, OAuth2, rate limiting, CORS
- Containerization: Docker + Docker Compose for easy deployment
- Monitoring: Sentry error tracking, OpenTelemetry tracing
- Documentation: Auto-generated API docs with FastAPI
- Testing: Pytest for backend, Jest + Playwright for frontend
- Docker & Docker Compose
- Python 3.10+ (for local development)
- Node.js 18+ (for local development)
git clone https://github.com/yourusername/Agent Builder.git
cd Agent Builder
# Copy environment configuration
cp .env.example .envEdit .env file with your API keys:
# LLM Providers (choose one or more)
OPENAI_API_KEY=your_openai_key
ANTHROPIC_API_KEY=your_claude_key
GOOGLE_API_KEY=your_gemini_key
GROQ_API_KEY=your_groq_key
# Database Configuration
POSTGRES_DB=Agent Builder
POSTGRES_USER=postgres
POSTGRES_PASSWORD=your_secure_password
# Optional: Advanced Features
REDIS_URL=redis://localhost:6379
MILVUS_HOST=localhost
MILVUS_PORT=19530# Start all services
docker-compose up -d
# Check service status
docker-compose ps- Frontend: http://localhost:3000
- API Documentation: http://localhost:8000/docs
- Admin Dashboard: http://localhost:3000/admin
Want to access from your phone or tablet? Run the setup script:
Windows:
setup-local-network.batMac/Linux:
chmod +x setup-local-network.sh
./setup-local-network.shThis will:
- β Detect your local IP address automatically
- β Configure frontend to use your IP
- β Show access URLs for other devices
Then access from any device on the same WiFi:
- Frontend:
http://YOUR_IP:3000 - Backend:
http://YOUR_IP:8000
π Detailed Guide: See LOCAL_NETWORK_ACCESS_GUIDE.md or νκ΅μ΄ κ°μ΄λ
- Open the workflow builder at http://localhost:3000
- Drag and drop blocks from the palette
- Connect blocks to create your workflow
- Configure each block with your parameters
- Click "Run" to execute your workflow
- Monitor execution in real-time
- Customer Support: Intelligent ticket routing and automated responses
- Content Creation: Multi-agent content generation and review workflows
- Data Processing: Automated data extraction, transformation, and analysis
- Report Generation: Scheduled reports with AI-powered insights
- Literature Review: Automated research paper analysis and summarization
- Experiment Design: AI-assisted experimental planning and execution
- Data Analysis: Multi-modal data processing and visualization
- Knowledge Discovery: Automated pattern recognition and insight generation
- Content Production: Multi-agent creative workflows for marketing
- Design Automation: AI-powered design generation and optimization
- Media Processing: Automated video, audio, and image processing
- Personalization: Dynamic content adaptation for different audiences
- Visual workflow builder with 50+ blocks
- 17 orchestration patterns including 2025 trends
- Multi-LLM support with intelligent routing
- Real-time execution monitoring
- Enterprise RAG with vector search
- Marketplace: Community plugin marketplace
- Collaboration: Real-time collaborative editing
- Templates: Pre-built workflow templates library
- API Gateway: Advanced API management features
- AI Assistant: Natural language workflow creation
- Auto-optimization: ML-powered workflow optimization
- Enterprise SSO: SAML, LDAP, and enterprise authentication
- Multi-tenant: SaaS-ready multi-tenancy support
- Edge Computing: Edge deployment capabilities
We welcome contributions from the community! Here's how you can help:
Found a bug? Please create an issue with:
- Clear description of the problem
- Steps to reproduce
- Expected vs actual behavior
- Environment details
Have an idea? We'd love to hear it! Please include:
- Use case description
- Proposed solution
- Alternative approaches considered
- Mockups or examples (if applicable)
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Make your changes with tests
- Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
Help improve our documentation:
- Fix typos and grammar
- Add examples and tutorials
- Translate to other languages
- Create video tutorials
- [Discord Server](https://discord.gg/Agent Builder) - Real-time chat and support
- [GitHub Discussions](https://github.com/yourusername/Agent Builder/discussions) - Feature requests and Q&A
- [Reddit Community](https://reddit.com/r/Agent Builder) - Share workflows and tips
- Community Support: GitHub Issues and Discord
- Enterprise Support: Contact us at enterprise@Agent Builder.com
- Security Issues: security@Agent Builder.com
- β GitHub Stars: Growing daily!
- π΄ Forks: Active community contributions
- π¦ Docker Pulls: Thousands of deployments
- π₯ Community: Developers from 50+ countries
This project is licensed under the MIT License - see the LICENSE file for details.
- β Free for commercial use
- β No vendor lock-in
- β Full source code access
- β Community-driven development
- β Enterprise-friendly license
Special thanks to:
- LangChain,LangGraph Team for the amazing AI framework
- React,ReactFlow Team for the visual workflow foundation
- FastAPI Team for the high-performance backend framework
- Open Source Community for continuous inspiration and contributions
