Agents

What the word agent means once the marketing is removed: a loop, some tools, and the permissions you decided to grant.

11 concepts, about 66 minutes of reading at roughly six minutes each. Every concept in this track needs a paid plan.

What is in this track

What an agent actually is — A chatbot stops when it has answered. An agent stops when the job is done. That gap is the whole category.
The loop: observe, decide, act — One move, repeated: look at what happened, choose the next action, do it, look again.
ReAct: think, then act — Reasoning and actions interleaved, so each choice is made after seeing what the last one did.
What a tool call really is — The model never runs your code. It writes a request, and your program decides whether to honour it.
Sub-agents and the clean context — A fresh, short context beats a long one. Sub-agents exist to protect the parent from the mess.
MCP: one plug for every tool — A protocol so a tool gets written once instead of once per product. Hosts, clients, servers, JSON-RPC.
Tool descriptions are untrusted input — The text describing a tool was written by whoever wrote the tool, and the model reads it as instructions.
Skills and progressive disclosure — A folder the agent reads when it becomes relevant, instead of a handbook pasted into every request.
Computer use — Screenshot in, click out. The fallback for software that has no API, and the most brittle tool an agent has.
Human in the loop — Oversight is worth what the person can actually check. Attach it to irreversible actions, with real arguments shown.
Long-horizon autonomy — Long runs fail by drift and repetition rather than by one dramatic wrong decision.