A Tour of Agents

Free · Open source · Runs in your browser

Build an AI Agent in
60 Lines of Python

Write the code. Run it in your browser. See exactly what LangChain, CrewAI, and AutoGen do under the hood.

9 lessons·~30 minutes·No signup required·MIT licensed

This is the entire agent loop

while True:
    response = llm(messages, tools)

    if response.tool_call:
        result = run_tool(response.tool_call)
        messages.append(result)
    else:
        break
User
LLM
Tool
Result
LLM
Done

Watch the loop run. This is what every agent framework does.

>>

Runs in Your Browser

Python via Pyodide. Real code, real API calls, real traces. No terminal, no venv, no Docker.

{}

No Framework Required

Raw HTTP calls to the LLM API. See the actual POST request. No SDKs hiding the work.

==

See What Frameworks Do

Every lesson shows the LangChain or CrewAI equivalent — then the plain Python underneath.

9 lessons. Each one builds on the last.

Start anywhere. Every lesson runs live in your browser.

From a single function call to a complete agent framework — learn tool calling, conversation memory, state management, policy gates, and self-scheduling.

One function. One HTTP POST.

That's where it starts. The first lesson takes 2 minutes.

Start Lesson 1 →