The rapid evolution of artificial intelligence has catalyzed a paradigm shift from isolated AI agents to collaborative multi-agent systems, enabling its applications in domains. The absence of standardized technical frameworks for AI agent interoperability leads to significant challenges, including high integration costs, limited reusability, and barriers to cross-system interconnection, impeding the scalability and trustworthiness of AI agent ecosystems. To address these challenges and enable an open, interconnected, and secure multi-agent ecosystem, a globally recognized technical standard is needed to define requirements for core components, interoperability, and application-level interactions among AI agents.