How sudoskys/openaibot works
Main System Components
The system should be represented as a microservices architecture with these key components:
a) Message Processing Services:
• Sender Service (start_sender.py)
• Receiver Service (start_receiver.py)
• Message Queue (RabbitMQ) as the central communication hub
b) Platform Adapters:
• Telegram
• Discord
• Kook
• Slack
Each with their specific event handlers and HTTP clients
c) Core Components:
• LLM Integration (OpenAI compatible)
• Plugin System
• Authentication System
• Cache System
• Memory Management
• Task Management
Component Relationships
[Platform Clients] → [Receiver Service] → [Message Queue] → [Sender Service] → [Platform Clients]
• ↕
• [Core Processing Layer]
Detailed Architecture Elements
a) External Services (Hexagons):
• RabbitMQ (Message Queue)
• OpenAI API (or compatible endpoints)
• Platform APIs (Telegram, Discord, etc.)
b) Internal Components (Rectangles):
• App Components (/app directory)
• LLMKira Core (/llmkira directory)
• Plugin System (/llmkira/extra/plugins)
• Cache Stores (Redis/Local)
• Memory Management
• Document Management
Key Interactions to Highlight
• Message flow between platforms and services
• Plugin execution paths
• Authentication flows
• Tool call processing
Important Notes
Annotations for:
• Required Python version (>=3.9)
• Supported platforms
• Authentication methods
• Plugin system capabilities
• Hook system integration points
Keep the abstraction level high enough to be understandable but detailed enough to show the sophisticated plugin and hook systems that make this project unique.