AgentHub: A Research Agenda for Agent Sharing Infrastructure

2025-10-03

AI Summary

Current LLM agent ecosystems suffer from fundamental challenges: poor discoverability, absence of standardized evaluation criteria, and fragmented governance. Existing efforts typically address isolated dimensions—such as distribution or naming—lacking a holistic infrastructure. This paper introduces the AgentHub research agenda, the first systematic framework integrating capability specification, trustworthy governance, and collaborative ecosystem design from a full software engineering perspective. Inspired by npm and Hugging Face, AgentHub incorporates protocol negotiation, semantic naming, and a distributed architecture to ensure capability clarity, transparent lifecycle management, cross-platform interoperability, secure governance, and seamless workflow integration. It establishes the first scalable, trustworthy, and composable infrastructure blueprint and technical pathway for open-source agent ecosystems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Agent Communication
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

LLM-based agents are rapidly proliferating, yet the infrastructure for discovering, evaluating, and governing them remains fragmented compared to mature ecosystems like software package registries (e.g., npm) and model hubs (e.g., Hugging Face). Recent research and engineering works have begun to consider the requisite infrastructure, but so far they focus narrowly -- on distribution, naming, or protocol negotiation. However, considering broader software engineering requirements would improve open-source distribution and ease reuse. We therefore propose AgentHub, a research agenda for agent sharing. By framing the key challenges of capability clarity, lifecycle transparency, interoperability, governance, security, and workflow integration, AgentHub charts a community-wide agenda for building reliable and scalable agent ecosystems. Our vision is a future where agents can be shared, trusted, and composed as seamlessly as today's software libraries.

Problem

• Addressing fragmented infrastructure for agent discovery and governance
• Establishing standards for agent capability clarity and interoperability
• Enabling seamless sharing and composition of LLM-based agents

Innovation

• Proposes AgentHub for agent sharing infrastructure
• Frames challenges for reliable agent ecosystems
• Enables seamless agent composition like software libraries