Foundations — what an AI agent is

In one line

An agent runs a loop: think, pick a tool, run it, look at the result, repeat until done.

Key idea

A chatbot answers once. An agent works in steps. Four parts make it work:

• Model — the brain that decides.
• Tools — actions it can take.
• Memory — what it remembers.
• Harness — the scaffolding that keeps it on track.

Smallest agent

from langchain.agents import create_agent
agent = create_agent(model="claude-sonnet-4-6", tools=[], system_prompt="Be helpful.")
agent.invoke({"messages": [{"role":"user","content":"Hi"}]})

Links

• Agents docs
• Full guide — build your first agent

My notes