Inside one turn: reason, then act
The idea: Each turn it writes a private thought, picks ONE action (a tool call), then reads the result — interleaving reasoning and acting (the ReAct pattern) instead of answering in one shot.
What you'll be able to do: You can explain the ReAct loop: reason → act → observe each turn, and why agents 'think' before each tool call.
The problem it solves: You've seen the loop turn. But what does the agent actually do on a single turn?
Builds on: The harness: the loop, made real
← Token awareness & the economics · Next: The self-correction loop: test, fix, repeat →
All lessons