ReAct (Reasoning + Acting)

Interleave reasoning with tool actions in a loop: thought → action → observation.

ReAct is a prompting pattern where the model alternates between reasoning and actions, using observations from tools to drive the next step.

Core Loop

  1. Thought: decide what to do next.
  2. Action: call a tool (search, code, DB, etc.).
  3. Observation: read the tool result.
  4. Repeat until a stopping condition.

Why It Matters for Agents

ReAct is the default mental model for tool-using agents: planning is not done once upfront; it is continuously updated based on real-world feedback.

Notes

  • Many frameworks implement a ReAct-like loop even when they do not explicitly label it as such.
  • In 2025+ models increasingly internalize ReAct behavior, but explicit prompting still matters for reliability.

Further Reading