AI Legion provides a framework for creating and managing multiple autonomous LLM agents that collaborate to achieve complex tasks. It is designed for developers and researchers experimenting with multi-agent systems and AI coordination, offering a console-based interface for interaction and debugging.

How It Works

The platform leverages LLMs (GPT-3.5-turbo or GPT-4) to power individual agents. These agents maintain their own state, including memory, goals, and notes, stored locally. They communicate and coordinate through a console interface, allowing users to send messages to all active agents. The system supports web search capabilities via Google Custom Search API integration.

Quick Start & Requirements

• Install: npm install
• Requirements: Node 10+, OpenAI API Key, Google Custom Search Engine ID, Google API Key. Google Custom Search API must be enabled in your Google Cloud account.
• Running: npm run start [# of agents] [gpt-3.5-turbo|gpt-4]
• Documentation: https://github.com/eumemic/ai-legion

Highlighted Details

• Supports multiple autonomous agents working collaboratively.
• Agents maintain persistent state (memory, goals, notes) for continuity.
• Console interaction allows direct messaging to all agents.
• Debugging facilitated by selective state deletion and replaying agent actions.
• GPT-4 support is available for enhanced agent capabilities.

Maintenance & Community

The project appears to be maintained by a single developer, eumemic. There are no explicit links to community channels or roadmaps provided in the README.

Licensing & Compatibility

The README does not specify a license. This may pose a restriction for commercial use or integration into closed-source projects.

Limitations & Caveats

GPT-3.5-turbo agents are prone to error loops that can rapidly consume API tokens. Agents may initially exhibit errors related to action parameter usage or formatting, requiring a learning period. Continuous monitoring is advised due to the risk of infinite loops.