AgentRob: From Virtual Forum Agents to Hijacked Physical Robots

2026-02-14

AI Summary

This work proposes a novel three-layer framework that enables community-driven, language-mediated coordination between multiple large language model (LLM)-based agents and physical robots—a capability largely absent in current systems that rely on direct control. By leveraging an online forum as the interaction medium, agents interpret user instructions via @-mention mechanisms, then orchestrate Unitree Go2/G1 robots through vision-language model (VLM)-powered controllers integrated with the Model Context Protocol (MCP). The architecture supports asynchronous, persistent, and identity-aware cross-domain interactions, allowing multiple embodied agents to operate concurrently within a unified forum environment. Experimental results demonstrate the feasibility and effectiveness of this paradigm, marking the first realization of a multi-agent, multi-robot collaborative system coordinated entirely through natural language in a public, community-oriented setting.

Technology Category

• Multiagent Systems: Coordination and Collaboration
• Intelligent Robots: Multi-Robot Systems
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Large Language Model (LLM)-powered autonomous agents have demonstrated significant capabilities in virtual environments, yet their integration with the physical world remains narrowly confined to direct control interfaces. We present AgentRob, a framework that bridges online community forums, LLM-powered agents, and physical robots through the Model Context Protocol (MCP). AgentRob enables a novel paradigm where autonomous agents participate in online forums--reading posts, extracting natural language commands, dispatching physical robot actions, and reporting results back to the community. The system comprises three layers: a Forum Layer providing asynchronous, persistent, multi-agent interaction; an Agent Layer with forum agents that poll for @mention-targeted commands; and a Robot Layer with VLM-driven controllers and Unitree Go2/G1 hardware that translate commands into robot primitives via iterative tool calling. The framework supports multiple concurrent agents with distinct identities and physical embodiments coexisting in the same forum, establishing the feasibility of forum-mediated multi-agent robot orchestration.

Problem

• LLM-powered agents
• physical robots
• online forums
• multi-agent orchestration
• autonomous agent integration

Innovation

• Model Context Protocol
• LLM-powered agents
• forum-mediated orchestration
• VLM-driven robot control
• multi-agent physical embodiment