AgentOrchestra: Orchestrating Multi-Agent Intelligence with the Tool-Environment-Agent(TEA) Protocol

TL;DR

This paper introduces the TEA protocol and AgentOrchestra framework, enabling more flexible, traceable, and adaptable multi-agent systems with hierarchical orchestration and continual self-evolution, leading to state-of-the-art performance.

Contribution

The paper presents the TEA protocol for unified environment, agent, and tool management, and the AgentOrchestra framework for hierarchical multi-agent orchestration with continual adaptation.

Findings

Achieved 89.04% on GAIA benchmark, outperforming baselines.

TEA improves traceability, reproducibility, and scalability in multi-agent systems.

AgentOrchestra demonstrates effective online tool refinement and dynamic instantiation.

Abstract

Recent advances in LLM-based agent systems have shown promise in tackling complex, long-horizon tasks. However, existing LLM-based agentprotocols (e.g., A2A and MCP) under-specify cross-entity lifecycle and context management, version tracking, and ad-hoc environment integration, which in turn encourages fixed, monolithic agent compositions and brittle glue code. To address these limitations, we introduce the Tool-Environment-Agent (TEA) protocol, a unified abstraction that models environments, agents, and tools as first-class resources with explicit lifecycles and versioned interfaces. TEA provides a principled foundation for end-to-end lifecycle and version management, and for associating each run with its context and outputs across components, improving traceability and reproducibility. Moreover, TEA enables continual self-evolution of agent-associated components through a closed feedback loop, producing improved versions while supporting version selection and rollback. Building on TEA, we present AgentOrchestra, a hierarchical multi-agent framework in which a central planner orchestrates specialized sub-agents for web navigation, data analysis, and file operations, and supports continual adaptation by dynamically instantiating, retrieving, and refining tools online during execution. We evaluate AgentOrchestra on three challenging benchmarks, where it consistently outperforms strong baselines and achieves 89.04% on GAIA, establishing state-of-the-art performance to the best of our knowledge. Overall, our results provide evidence that TEA and hierarchical orchestration improve scalability and generality in multi-agent systems.