Flashcard: what is the ReAct pattern for agents, in one sentence?

Attempt it

ReAct (Reason and Act) is a pattern where the LLM alternates between an explicit Thought (a short reasoning step in natural language), an Action (typically a tool call), and an Observation (the real result of that action), repeating the cycle until it emits a Final Answer. The interleaving is what turns it from a single forward pass into an agent loop: each Observation enters the context before the next Thought, so the next reasoning step is conditioned on real data rather than the model's own continuation. Introduced by Yao et al. in 2022 (arXiv 2210.03629), ReAct became the default substrate for tool-using agents because the closed loop bounds hallucination drift on multi-step tasks. Modern frameworks (LangGraph, CrewAI, AutoGen) and provider-native loops (Claude Opus 4.7 computer use, GPT-5.5 tool calling) all use a ReAct-shaped trajectory under the hood.

Your answer

Quick reference

TL;DR

ReAct alternates Thought, Action, and Observation in a loop, so each reasoning step is conditioned on a real tool result rather than the model's own continuation.

Memory aid

Sign in for the mnemonic that makes this stick. See Pro pricing.

Easy to grasp

Imagine a chef cooking a new dish. One way is to read the whole recipe, then cook the entire meal from memory without looking again. Another way is to read one step, do it, taste the result, then decide what to do next based on what actually happened. ReAct is the second way for an LLM. The model thinks a little, takes one action like searching the web or calling a calculator, looks at the real result, and only then thinks about what to do next. The thinking and the doing take turns, and each new thought stands on something the model actually observed instead of something it guessed. That back-and-forth is what turns one model call into an agent loop.

Key concepts

ReAct Pattern Agent Loop Tool Use Chain-of-thought

Detailed answer & concept explanation

~7 min read

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example. Click to expand.

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example.

Sign in for 5 free unlocks/day or go Pro for unlimited. See Pro pricing.

Quick scan · 2–4 min

Everything important, quickly.

Name the three steps in order: Thought, Action, Observation. Explain that the Observation is written by the runtime, not the model, which is what makes it external evidence. Contrast with chain-of-thought as open loop versus ReAct as closed loop. Attribute to Yao et al. 2022. Close with the tradeoff: extra latency and context growth.

Real-world examples

Real products, models, and research that use this idea.

LangGraph's prebuilt ReAct agent wraps Anthropic or OpenAI function calling in a state graph where each node is one Thought, Action, Observation turn with full trace logging.
Claude Opus 4.7 in computer-use mode runs a ReAct-style loop: it observes a screenshot, reasons, emits a click or keystroke, then observes the next screenshot before the next Thought.
Perplexity research mode interleaves reasoning with live web searches, so each answer is grounded in retrieved sources rather than the model continuing from its parametric memory.
Cursor and Cline coding agents observe file diffs and test output between edits, re-grounding the next reasoning step on real results instead of a guessed code state.

Interviewer follow-ups

What an interviewer would ask next. Try answering before peeking at the approach.

QHow does ReAct differ from plan-and-execute, in one sentence?

Sign in or go Pro to see the approach for this follow-up.

QWhat bounds hallucination drift in a ReAct loop, and what reintroduces it?

Sign in or go Pro to see the approach for this follow-up.

QWhy is provider-native function calling a natural fit for ReAct?

Sign in or go Pro to see the approach for this follow-up.

Don't say this

Red flags and common mistakes that signal junior thinking. Click to expand.

Most common mistake

Describing ReAct as just chain-of-thought plus tools. The defining feature is the interleaving order: Observation enters the context before the next Thought.

Sign in or go Pro to see all red flags and common mistakes. See Pro pricing.

60-second cheat sheet

The night-before-the-interview bullets. Scan these on the way to the call.

Sign in for 5 unlocks/day or go Pro for unlimited revision sheets. See Pro pricing.

References

Primary sources. Skim if you want the original framing.

Yao et al., ReAct: Synergizing Reasoning and Acting in Language Models