Skip to content

levilevente/AEGIS

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Project A.E.G.I.S.

Autonomous Executive General Intelligence System

A.E.G.I.S. is an experimental agentic framework designed to bridge the gap between LLMs and functional personal assistants (like Jarvis). It is not just a chat interface; it is a state-aware system capable of visual perception, decision-making, and tool orchestration.

Powered by LangGraph, this project moves away from linear chains and towards cyclical, graph-based reasoning. The ultimate goal is to create a Domestic Intelligence Hub—a system that can debug your code and manage your smart home environment seamlessly.

⚡ Core Capabilities

  • 🌐 Graph-Based Orchestration: Utilizing LangGraph to manage agent state, allowing for loops, retries, and conditional logic.
  • 🏠 Smart Home Nexus: Designed to integrate with home automation APIs (e.g., Home Assistant, Philips Hue) to control the physical environment.
  • 👁️ Visual Cortex: Integrated multimodal capabilities (GPT-4o / Claude 3.5 Sonnet) to analyze real-time video feeds or uploaded snapshots.
  • 🖥️ Command Center UI: A futuristic, functional dashboard built with Streamlit to visualize the agent's "thought process" and outputs.
  • 🛠️ Extensible Tooling: A modular design allowing the agent to connect to local files, web search, and system APIs.

About

Project A.E.G.I.S. is an open-source initiative to build a personal intelligent system reminiscent of pop-culture assistants like Jarvis or F.R.I.D.A.Y. Built on a robust LangGraph architecture, it moves beyond simple "chatbot" interactions to a graph-based workflow capable of state management, visual reasoning, and tool execution.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors