C.W.K.
Stream
Lesson 03 of 05 · published

ReAct and the Agent Landscape

~34 min · react, frameworks, landscape

Level 0Observer
0 XP0/40 lessons0/12 achievements
0/100 XP to next level100 XP to go0% complete

ReAct is the old root that still matters

ReAct means reasoning plus acting: the model reasons about the next move, calls a tool, observes the result, and reasons again. Modern APIs hide some of the scratchpad, and good systems should not leak private reasoning to users, but the control pattern is still everywhere.

The practical lesson is not "show chain-of-thought." The practical lesson is "alternate decisions with evidence." A tool call without an observation is wishful acting. A long internal monologue without tools is just expensive prose.

Do not confuse products with primitives

Frameworks differ in surface area, but most are combinations of the same primitives: model call, tool schema, tool executor, state, memory, handoff, guardrail, trace, and stop condition. Learn the primitives first. Then choose the product.

The 2026 decision map

Use a raw loop when the system is small and control matters. Use OpenAI Agents SDK when you want code-level orchestration, hosted OpenAI tools, handoffs, guardrails, and tracing. Use LangGraph when state transitions, persistence, replay, and human interrupts are the product. Use CrewAI when role/task configuration is the main value. Use Claude Agent SDK when a Claude-native developer harness, files, shell, MCP, and permissioned long-running sessions are central.

Code

ReAct as a private control trace·text
Goal: answer a user question with current evidence
Decision: need fresh source
Action: web_search("OpenAI Agents SDK tracing")
Observation: official docs say tracing records model calls, tools, handoffs, guardrails
Decision: read official docs, then answer with citation
Action: fetch_url(official_docs_url)
Observation: exact section found
Final: answer with source and caveat
Framework decision table·text
Small single-agent tool loop      -> raw API loop
OpenAI tools + handoffs + traces   -> OpenAI Agents SDK
Durable graph state + replay       -> LangGraph
Role/task team simulation          -> CrewAI
Developer workspace harness        -> Claude Agent SDK
Data-heavy retrieval agent         -> LlamaIndex
Quick code-as-action prototype     -> Smolagents

External links

Exercise

Pick one agent idea and classify it twice: first by ReAct primitives, then by framework fit. Name the primitive that justifies the framework.
Hint
If you cannot name the primitive, choose the raw loop first.

Progress

Progress is local-only — sign in to sync across devices.
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.