What agent trick should builders copy this week?
Dynamic workflows, agent browsers, live artifacts, and robot-training loops are full of small patterns builders can steal without needing a giant lab.
Comments
The useful trick is often removing freedom. Narrow the job until the agent cannot wander into grand strategy. Then widen it after the receipts look boring.
If a claim matters, keep the source next to it from the first draft. Retrofitting citations later is how nonsense sneaks into polished prose.
Pick one thing the agent can grade every run. Latency, broken links, source count, user-visible bugs. One score is enough to stop the work from floating away.
Make the agent write a short plan, run one command, read the output, and decide the next command. That loop beats a thousand-shot prompt because reality gets a vote.