PolyAgents

PolyAgents is an advanced multi-agent system leveraging the Google Gemini API, designed to enable intelligent conversations through a sophisticated consensus mechanism. This project emphasizes real-time streaming capabilities, heightened security, comprehensive health monitoring, and effective error handling.

🚀 Features

Core Features

• Multi-Agent Conversations: Seamlessly orchestrates interactions among multiple Gemini agents, each configured with varying models and temperatures.
• Intelligent Consensus Mechanism: Utilizes LLM-powered synthesis to merge insights from various agents for a well-rounded perspective.
• Real-Time Streaming: Incorporates WebSocket support to provide live updates during conversations.
• Memory Layer: Utilizes Redis Streams for inter-agent communication along with PostgreSQL for data persistence.
• Vector Search: Optional integration with Qdrant to enhance semantic search capabilities.
• Comprehensive Agent Responses: Allows for full-length responses through expandable modal views, ensuring complete information delivery.
• Progress Tracking: Offers real-time status updates throughout the agent processing lifecycle.

Security & Authentication

• API Key Authentication: Ensures secure access control with customizable permissions.
• Rate Limiting: Implements sophisticated rate limiting with per-client tracking to manage bursts.
• JWT Token Support: Provides stateless authentication methods with configurable token expiration.
• Input Validation: Ensures thorough sanitization and validation of all inputs to mitigate risks.
• IP Blocking: Automatically restricts access from malicious IPs.
• CORS Configuration: Flexible cross-origin resource sharing settings to enhance security.

Monitoring & Observability

• Advanced Health Checks: Monitors system health at component levels, tracking response times.
• System Metrics: Offers real-time statistics related to conversations, agent performance, and errors.
• Circuit Breaker Pattern: Incorporates automatic fault tolerance for external services.
• Error Tracking: Detailed logging of errors, providing contextual insights for troubleshooting.
• Performance Monitoring: Monitors response times and identifies potential bottlenecks.

Resilience & Reliability

• Retry Logic: Employs exponential backoff with jitter for managing failed operations.
• Graceful Degradation: Manages fallback responses seamlessly when services are down.
• Connection Pooling: Enhances resource management efficiency for database interactions.
• Background Tasks: Automates cleanup and maintenance operations to optimize performance.
• Service Discovery: Implements health-aware service routing and load balancing.

📋 Requirements

• Python 3.8+
• Google Gemini API key
• Redis 6.0+
• PostgreSQL 12+
• Qdrant (optional)

🎨 Frontend

The PolyAgents frontend is constructed using:

• React 18 with TypeScript
• Vite for fast development
• Tailwind CSS for modern styling
• Shadcn/ui for UI components
• WebSocket for real-time capabilities

Frontend Features

• Real-time chat interface enabling live conversations with multiple agents.
• Agent status panel for monitoring agent activity and system health.
• Conversation management tools for browsing, searching, and managing interactions.
• Responsive design crafted for both desktop and mobile devices, ensuring a versatile user experience.
• Dark mode support, optimized for extended use.

🌐 API Endpoints

Robust API endpoints facilitate various functionalities, including secure authentication and real-time chat interactions.

Authentication

Protected endpoints require an API key for access:

Authorization: Bearer your_api_key_here

Chat Endpoints

POST /chat

Initiates a conversation among agents, aggregating their insights.

Request example:

{
  "message": "What are the benefits of renewable energy?",
  "conversation_id": "optional-uuid",
  "agents": {
    "count": 3,
    "turns": 2,
    "temperature": 0.7,
    "model": "gemini-pro"
  }
}

WebSocket Interface

Utilizes a WebSocket connection to deliver real-time updates on conversation status and agent responses.

Health Monitoring

Includes comprehensive health checks for system components, ensuring reliable operation and performance.

This system's structured architecture and modular features contribute significantly to the development of intelligent, responsive agents and robust conversation management.