The agent loop is just ReAct, and your tool-use API already implements it
Summary
The article argues that the agent loop is fundamentally the ReAct pattern, and that current tool-use APIs already implement this mechanism.
Similar Articles
@irl_danB: everyone is building an agent or a tool you don't want an agent or a tool, you want a reactor I've been working on some…
A developer introduces a concept called a 'reactor' — an agent session DAG that maintains a memoized world-model using OpenProse markdown files and the openai-agents-sdk, drawing parallels to React and dataflow.
Why are so many agent tools just 1:1 API wrappers?
The article argues that many AI agent tools are just 1:1 API wrappers, pushing branching logic into the LLM and causing failures. The author recommends task-shaped tools like upsert_contact that encapsulate search/create/update logic in code, pass known context, validate inputs, and return structured errors.
Open catalog of agent patterns + the frameworks that implement them
An open catalog of agent patterns (e.g., ReAct, tool-use, RAG) and their implementations across frameworks like LangChain, LlamaIndex, and others, aiming to map the agent landscape using a pattern language approach.
ReAct or CodeAct, that is the question
The article discusses the trade-offs between ReAct and CodeAct orchestration paradigms in AI engineering, highlighting CodeAct's efficiency for complex tasks and introducing a new open-source framework.
AI agents are useful, but agent loops still make me nervous
The article discusses the usefulness of AI agents while expressing concerns about the potential risks of agent loops.