atelier-grill-me
Grill me
Pressure-test a plan or design by interviewing the user until there is genuinely shared understanding: every consequential branch of the decision tree resolved, and the dependencies between those decisions settled in order. The point is not interrogation for its own sake: it is to surface the assumptions, edge cases, and forks now, at conversation cost, instead of mid-build or in production.
This is the on-demand, relentless counterpart to "think before coding." That habit asks for clarification when warranted; this skill commits to walking the entire tree before a line is written.
Interaction and agent discipline: as the main atelier skill's Interaction section (terse, answer first, no em dashes, each question led by your recommendation, next steps when the interview wraps), including its two gates on commits and tests (rules 24-25); the one-round cap does not apply here, multi-round questioning is this skill's sanctioned purpose.
When to use
- The user asks to be grilled, or to stress-test / pressure-test a plan or design.
- A decision carries real stakes and a bad call is expensive to reverse: architecture, data model, public API shape, a migration, a dependency choice, a security boundary.
- The plan is a new product or feature: before grilling how to build it, grill whether to build it at all (the atelier
references/product.md§ Validate before you build). What evidence of the problem exists beyond the room? What is the cheapest test of demand (a landing page, a concierge run) before the build? What dated go/no-go criteria would make "no" sayable? What adoption threshold decides keep-or-kill after launch? A killed idea at interview cost is this skill's best outcome. - A plan is vague, broad, or hides many unstated branches.
Match intensity to stakes. A five-question interrogation of "rename this variable" is noise: skip the grilling for trivial or already-well-specified tasks.