# Grilling > Stress-test a plan, decision, or idea through relentless questions and explicit grill requests. - Skill: `jesse-merhi/grilling` (Agent Skill) - Install (CLI): `npx skillmds@latest add jesse-merhi/grilling` - Raw SKILL.md: https://api.skillmd.com/api/skills/jesse-merhi/grilling/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - Author: jesse-merhi (https://skillmd.com/u/jesse-merhi) - Updated: 2026-09-22 - Page: https://skillmd.com/skills/jesse-merhi/grilling --- # Grilling Interview the user relentlessly until you reach a shared understanding. Map the subject 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. Ask the questions you can ask now without guessing at answers you have not heard yet. Ask the whole frontier in one numbered round and give your recommended answer for every question. Then wait for the user's answers before continuing. Format each round with a horizontal rule between questions so the recommendations do not run together: ```markdown ❓ **Q1**. ****: ➡️ --- ❓ **Q2**. ****: ➡️ ``` Each answer reshapes the tree. Recompute the frontier after every round. A question that depends on another question still open in this round belongs to a later round. Finding facts is your job. When a frontier question needs a fact from the environment, dispatch a background subagent to find it. Keep asking the rest of the frontier while that exploration runs; only questions downstream of the unsettled fact wait. Decisions belong to the user. Put each one to them and wait for their answer. The session is done when the frontier is empty. Every branch has been visited and nothing is left silently assumed. Do not act on the result until the user confirms you have reached a shared understanding.