All It Takes Is One Prompt: An Autonomous LLM-MA System

Qian Wang, Tianyu Wang, Zhenheng Tang, Qinbin Li, Nuo Chen, Jingsheng Liang, Bingsheng He

Keywords: Hierarchical Agent System

Abstract: LLM-based multi-agent (LLM-MA) systems have shown promise in tackling complex tasks. However, existing solutions often suffer from limited agent coordination and heavy reliance on predefined Standard Operating Procedures (SOPs), which demand extensive human input. To address these limitations, we propose \textit{MegaAgent}, a framework designed for autonomous coordination in LLM-MA systems. \textit{MegaAgent} generates agents based on task complexity and enables dynamic task decomposition, parallel execution, efficient communication, and comprehensive system monitoring of agents. In evaluations, \textit{MegaAgent} demonstrates exceptional performance, successfully developing a Gobang game within 800 seconds and scaling up to 590 agents in a national policy simulation to generate multi-domain policies. It significantly outperforms existing systems, such as MetaGPT, in both task completion efficiency and scalability. By eliminating the need for predefined SOPs, \textit{MegaAgent} demonstrates exceptional scalability and autonomy, setting a foundation for advancing true autonomy in LLM-MA systems. Code is available at \url{https://anonymous.4open.science/r/MegaAgent-dev-DEF0}

Submission Number: 62