LLM Agent

A software system built around a large language model that autonomously perceives state, plans actions, executes them against an environment (a web page, a mobile app, a shell, a UI), and reflects on outcomes to make progress toward a goal. In accessibility work, LLM agents are increasingly used to execute realistic user tasks through assistive technology proxies — for example, driving an Android app via a modified TalkBack service to surface errors that affect screen reader users. Multi-agent designs typically separate a Decision agent (proposes the next action), a Reflection agent (evaluates the outcome), and a shared Memory for carrying task-relevant context across steps.