01
Delegate repetition, keep judgement
Repetition is the signal to look for. The second time you write the same runbook, review comment or query, that task is a candidate for an agent.
02
Review the decision, not the keystrokes
Ask for the plan and the evidence before approving an action. Reading a plan takes a minute and teaches you how the agent reasons.
03
Send it back with a reason
When output is wrong, write down why. The reason narrows the next attempt; a silent rejection guarantees the same work comes back.
Week one: read-only
Agents observe and report. Compare their conclusions with yours to learn where they are reliable.
Then: propose and wait
Agents draft the change and stop. A named person approves before anything lands, so mistakes stay cheap.
Steady state: act in a boundary
Agents act inside a scope you wrote down, and every action leaves a trail you can read afterwards.
Why this matters
Time is saved where work stops waiting on a person, not where prompts get longer.