grillinglisted

Interview the user relentlessly until you reach a shared understanding. Map this as a **design tree**: every decision branches into the decisions that hang off it. Work the tree in **rounds**. The **frontier** is every decision whose prerequisites are already settled: the questions you can ask _now_ without guessing at answers you haven't heard yet. Ask at most four frontier questions in one round, starting with the ones whose answers remove the most other questions: number each question and give your recommended answer. Then wait for the user's answers before the next round. Every question offers the smallest option first: do nothing, reuse what exists, or defer. Recommend it unless you can name a concrete case the user will hit where it fails. When you recommend more, name what it adds (a field, a state, a command, a file, a concept) and what breaks without it. Format a round like so: ``` ❓ **Q1** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices> ➡️ <your recommended answer> --- ❓ **Q2** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices> ➡️ <your recommended answer> ``` When every question in a round has a short set of options, and your harness has a multiple-choice question tool (such as `AskUserQuestion` in Claude Code), ask the round through it instead, with the recommended option first. Each round the user answers reshapes the tree: settled decisions push the frontie