grilling

Interrogates plans and decisions through iterative question rounds mapped as a design tree.

Updated Sep 14, 2026
One-click install
npx skills add https://github.com/Suge8/skills --skill grilling-suge8
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: grilling
Source: https://github.com/Suge8/skills/tree/main/workflow/grilling
Command: npx skills add https://github.com/Suge8/skills --skill grilling-suge8

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Plans, decisions, and ideas often contain hidden assumptions and unresolved dependencies that surface too late. This Skill stress-tests your thinking by relentlessly questioning every decision until nothing is silently assumed. ## Core Features & Use Cases - Design Tree Mapping: Models your plan as a tree of decisions where each choice branches into dependent sub-decisions. - Frontier-Based Rounds: Each round asks only the questions whose prerequisites are already settled, with numbered questions and recommended answers. - Fact Lookup via Subagents: Dispatches subagents to gather environmental facts (filesystem, tools) instead of asking you for information it can find itself. - Use Case: Before starting a new feature, run a grilling session to expose every unstated assumption in your design, then proceed to ticket breakdown or implementation only after full alignment. ## Quick Start Ask the agent to grill you on your plan for the upcoming feature so every hidden assumption is questioned before implementation begins.

Frequently Asked Questions about grilling

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I stress-test a software design before implementation?▼

Use a structured interrogation approach that maps your plan as a design tree of dependent decisions. Each round asks only the questions whose prerequisites are settled, with recommended answers, until no assumptions remain unexamined.

What is a design tree for decision making?▼

A design tree models a plan where each decision branches into the decisions that depend on it. The frontier is the set of questions answerable now, and each answered round pushes the frontier outward to unblock dependent questions.

Can the agent look up facts instead of asking me?▼

Yes. When a question requires environmental facts like filesystem or tool state, the agent dispatches subagents to find them rather than asking you. Only genuine decisions are left for the user to make.

When does a grilling session end?▼

The session ends when the frontier is empty, meaning every branch of the design tree has been visited and nothing is silently assumed. The agent waits for your confirmation of shared understanding before taking any action.

When should I not use iterative design interrogation?▼

Avoid it for trivial tasks with no meaningful design decisions or when requirements are already fully specified and validated. It adds the most value for ambiguous plans with hidden dependencies and unstated assumptions.