Agents & tools · Glossary term
What is ReAct?
An agent pattern that interleaves task reasoning, a concrete action, and an observation returned by the environment before deciding the next step.
Why does ReAct matter?
Environment feedback can correct assumptions and ground later decisions instead of forcing the model to complete the entire task from internal generation alone.
ReAct in practice
Expose a small set of typed tools, return concise observations, cap the loop, and verify the final artifact rather than storing private reasoning traces.
What is the common confusion about ReAct?
ReAct is a prompting and control pattern, not a guarantee of autonomy, correctness, or safe tool use.
Learn ReAct in the course
Lessons that name ReAct in a title or section
- Few-Shot, Chain-of-Thought, Tree-of-Thought
Telling a model what to do is prompting. Showing it how to think is engineering. The gap between 78% and 91% accuracy on the same model, same task, same data is not a better model.
- Agent State Machines — Graphs, Nodes, Checkpoints
A ReAct loop written by hand is a while True. The same loop written as an explicit graph is something you can checkpoint, interrupt, branch, and time-travel through. The agent hasn't changed.
- The Agent Loop: Observe, Think, Act
Every agent in 2026 is a variant of the ReAct loop from 2022 — Claude Code, Cursor, Devin, Operator included. Reasoning tokens interleave with tool calls and observations until a stop condition fires.
Covered in Phase 11: LLM Engineering and Phase 14: Agent Engineering.
Related terms
- AgentA software system that lets a model select actions toward a goal, observe tool or environment results, and continue under an orchestration…
- Function CallingA provider or application interface through which a model emits a structured request naming a tool and its arguments.
- PlanningConstructing, selecting, or revising a sequence of actions and dependencies intended to move from the current state to a goal.
- GroundingConnecting a generated answer or action to evidence, state, or observations that the system can identify and check.
Sources
More terms in Agents & tools
- Agent
- Agent Harness
- Agent Memory
- Agent State
- Agent Skill
- Approval Gate
- Checkpoint
- Compensating Action
- Delegation
- Durable Execution
- Function Calling
- Human-in-the-Loop (HITL)
- MCP (Model Context Protocol)
- Multi Round-Trip Request (MRTR)
- Orchestration
- Planning
- Sandbox
- Skill Bundle
- Skill Catalog
- Skill Discovery
- Skill Invocation
- Stateless MCP
- Structured Output
- Swarm
- Termination Condition
- Tool Contract
This entry comes from glossary/terms.md on GitHub. Browse all 250 glossary terms.