plan-and-critique

A brainstorm and critique loop structures open-ended design work by widening the option space, drafting an artifact, and running adversarial self-review to catch hidden assumptions before finalizing architecture plans.

To avoid premature drafting in system design, generate a brief brainstorm to explore multiple approaches, ask high-leverage clarification questions, then critique the initial draft adversarially to enforce stopping rules before delivery.

Yes, the disciplined revision process applies to research plans and complex documents by requiring explicit assumptions, focused clarification, and adversarial self-review to prevent endless iteration and premature generalization.

The best way to critique a design plan is adversarial self-review, actively testing the artifact for hidden assumptions, weak coverage, and missing stopping criteria, then revising honestly while tracking unchanged elements.

You should not use an adversarial review cycle for trivial tasks with only one reasonable approach, as the overhead of explicit assumptions and structured critique provides no value where a weak first pass is low cost.