zejzl.net
Multi-Agent AI Orchestration Framework
An async message bus AI framework that orchestrates multiple AI models through a specialized 9-agent system for complex task decomposition, execution, validation, and continuous learning.
Real-time metrics and performance indicators from the zejzlAI framework.
Get zejzlAI running in under 5 minutes. Choose your preferred installation method:
One command, production-ready setup with all dependencies:
# Clone and deploy with Docker Compose git clone https://github.com/zejzl/zejzlAI.git cd zejzlAI/zejzl_net docker-compose up -d # Access web dashboard # Open: http://localhost:8000
Full control with Python virtual environment:
# 1. Clone repository git clone https://github.com/zejzl/zejzlAI.git cd zejzlAI/zejzl_net # 2. Install dependencies pip install -r requirements.txt # 3. Configure API keys (optional) cp .env.example .env # Edit .env with your API keys # 4. Run the framework python main.py
python ai_framework.py chat chatgpt "Explain quantum computing"
python main.py -mb --task "Design REST API for todo app"
python main.py --pantheon --task "Analyze codebase"
💬 Need help? Check the GitHub Issues or read the full documentation.
Explore the internal workings of the zejzlAI framework through interactive diagrams and system visualizations.
Complete system overview showing the 9-agent pantheon, message bus, and AI provider integration.
High-performance async message routing with 408K+ msg/sec throughput and priority queuing.
Observer → Reasoner → Actor → Validator flow with parallel execution and consensus detection.
Real-time collaboration between specialized agents with consensus detection and task distribution.
Throughput benchmarks, latency tracking, and system telemetry dashboard with real-time monitoring.
Circuit breakers, vitality boosts, and auto-recovery mechanisms for resilient operations.
┌─────────────────────────────────────────────────────────────┐
│ 🖥️ CLI Interface │
│ (Interactive Menu) │
├─────────────────────────────────────────────────────────────┤
│ 🔄 AsyncMessageBus Layer │
│ ┌────────────────────────────────────┐ │
│ │ • Provider Management │ │
│ │ • Message Routing (408K+ msg/sec) │ │
│ │ • Conversation History │ │
│ │ • Circuit Breakers │ │
│ └────────────────────────────────────┘ │
├─────────────────────────────────────────────────────────────┤
│ 🤖 9-Agent Pantheon │
│ │
│ 👁️ Observer ──→ 🧠 Reasoner ──→ ⚡ Actor ──→ ✅ Validator │
│ ↓ ↓ ↓ ↓ │
│ 🔧 Executor ←── 💾 Memory ←── 📊 Analyzer ←── 📚 Learner │
│ ↓ │
│ ✨ Improver │
│ (Magic System) │
├─────────────────────────────────────────────────────────────┤
│ 🌐 AI Provider Layer (7 Providers) │
│ ChatGPT │ Claude │ Gemini │ Grok │ DeepSeek │ Qwen │ Zai │
│ • Rate Limiting (60/min, 1000/hour) │
│ • Smart Retry with Exponential Backoff │
│ • Circuit Breakers & Auto-Recovery │
├─────────────────────────────────────────────────────────────┤
│ 💾 Persistence Layer (Hybrid Storage) │
│ Redis (Primary) + SQLite (Fallback) │
│ • Conversation History • Agent State │
│ • Performance Metrics • Security Audit │
└─────────────────────────────────────────────────────────────┘
Click any image above to view full-screen! 🔍
📚 Explore Full DocumentationEach agent has a unique personality with specialized communication styles and expertise areas, working together through an async message bus for complex AI orchestration.
Thorough task analysis with systematic decomposition and pattern recognition.
Strategic planning with risk assessment and dependency mapping.
Practical execution with efficient action planning and rapid deployment.
Rigorous quality assurance with detailed validation reports and testing.
Performs validated tasks with comprehensive error handling and recovery.
Stores and recalls execution state, history, and learned patterns.
Deep performance analysis with actionable insights and metrics tracking.
Identifies patterns, bottlenecks, and generates optimization strategies.
Innovative system optimization with magic-based self-healing capabilities.
High-performance async/await architecture with 408K+ msg/sec throughput and sub-millisecond (0.007ms) latency.
Integrate ChatGPT, Claude, Gemini, Grok, DeepSeek, Qwen, and custom AI models seamlessly.
Redis primary with SQLite fallback for reliable state management and data persistence.
Circuit breakers, auto-recovery, and vitality boosts for resilient operations.
End-to-end AES-256-GCM encryption, secure key management, and GDPR compliance.
Real-time telemetry, profiling, and comprehensive analytics dashboard.
Production-ready containerization with health checks and orchestration support.
Model Context Protocol support with 6+ built-in servers and 36+ tools.
Comprehensive guides, API references, and tutorials to get the most out of zejzlAI.
Can't find what you're looking for?