Agent Exchange: An Auction Platform for AI Agent Marketplaces

Yingxuan Yang, Ying Wen, Jun Wang, Weinan Zhang

Keywords: AI Agent, Agent-Centric Economy, Multi-Agent Systems, Auction Mechanism

Abstract: The rise of Large Language Models (LLMs) has transformed AI agents from passive tools into autonomous economic participants, marking the emergence of the agent-centric economy where agents exchange value, make strategic decisions, and coordinate with minimal human oversight. We propose Agent Exchange (AEX), an auction platform designed for AI agent marketplaces. Inspired by Real-Time Bidding (RTB) in online advertising, AEX serves as the central auction engine coordinating four ecosystem components: the User-Side Platform (USP) for translating human goals into agent-executable tasks; the Agent-Side Platform (ASP) for capability representation and performance tracking; Agent Hubs for team coordination and auction participation; and the Data Management Platform (DMP) for secure knowledge sharing and fair value attribution. We present AEX's design principles, system architecture, and empirical evaluation focusing on the core auction mechanisms. Our contribution lies in engineering a practical auction-based coordination framework and demonstrating that competitive multi-level selection improves task outcomes over direct assignment, laying the groundwork for agent-based economic infrastructures in future AI ecosystems.

Track: Long Paper

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 108