First-party only — a session id and where you arrived from, so we can tell which pages are worth writing more of. No advertising, no third-party trackers, nothing that follows you off this site.
Shorter pieces than the tutorials: what changed in a tool and whether it should change how you work, patterns that turned out to matter, and postmortems of agent-written code that went wrong.
Getting an AI coding agent to behave: config files, permissions, hooks, memory, and the boring switches that decide whether it helps or hurts.
BrowseHow to describe work so an agent gets it right the first time — specs, constraints, examples, and the failure modes behind each.
BrowseEnd-to-end ways of working: review loops, parallel agents, test-first pipelines, and where a human still has to stand.
BrowseWhat changed in Claude Code, Codex, Cursor and the rest — and whether it changes how you should work.
BrowseThings that went wrong with agent-written code, traced back to the decision that caused them.
BrowseThe categories above are where writing will land. In the meantime, the tutorial tracks are the substantial material on this site — the Foundations track is free and complete.