Meet Alibaba’s Page Agent: A JavaScript In-Page GUI Agent That Controls Web Interfaces With Natural Language Through the DOM

Alibaba researchers have introduced Page Agent, a tool that operates as client-side JavaScript to control web interfaces using natural language commands. By interacting directly with the Document Object Model, the system allows users to automate browser tasks without requiring backend modifications or visual processing. This approach simplifies the development of web-based AI agents by enabling direct text-based manipulation of page elements.