ahd94

I was working on some feature for tinyagents.dev and i kept updating the idea while working on it.

I wanted to build few features and then I changed my plan but in every new session it would totally misunderstand the context and i had to finally pause and ask it reindex and update its memory entirely.

This is especially worse when you have say openai and claude code working in tandem.

9 short lessons, ~60 lines of Python. Each adds one concept — tool calling, memory, state, policy gates, self-scheduling — until you see the full agent stack.

The core loop is just a while loop. That's AgentExecutor. That's what the frameworks do.

Runs in the browser (Pyodide). No setup, no signup. Mock mode works instantly, or plug in a free Groq API key for live LLM responses.

Open source: https://github.com/ahumblenerd/tour-of-agents