ML//agent//ReAct

ReAct interleaves reasoning with acting. The model uses current evidence to select a tool, observes its result, and revises the next decision instead of committing to an entire plan upfront.

ReAct interleaves reasoning with acting. The model uses current evidence to select a tool, observes its result, and revises the next decision instead of committing to an entire plan upfront.

Its important contribution is not a particular prompt format. It makes external feedback part of inference rather than an afterthought.

Tool errors and surprising observations become opportunities to update the trajectory, provided the harness returns them legibly.

Modern agents often hide or restructure reasoning traces, but the observable control pattern remains: propose, act, inspect, continue.