grill-me

Interrogate decision branches to clarify and validate plans or designs.

Updated May 3, 2026
One-click install
npx skills add https://github.com/tabesink/deepdoc-agent --skill grill-me-tabesink
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grill-me
Source: https://github.com/tabesink/deepdoc-agent/tree/main/.cursor/skills/grill-me
Command: npx skills add https://github.com/tabesink/deepdoc-agent --skill grill-me-tabesink

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes ambiguity from plans and designs by forcing explicit decisions, surfacing hidden assumptions, and reconciling conflicting branches until both you and the system share the same understanding.

Core Features & Use Cases

  • Relentless design interrogation: Breaks a plan into a decision tree and asks targeted questions for each branch.
  • Dependency-driven clarification: Resolves dependencies between decisions one-by-one to prevent downstream surprises.
  • Code-first verification: If a question can be answered by reading the codebase, it switches to exploring the codebase rather than guessing.

Quick Start

Ask the AI to grill you on your plan and explicitly resolve each open decision branch, e.g., "Grill me on my system design for recursive RAG retrieval and synthesis with citations, and keep drilling until we agree on every dependency and assumption."

Frequently Asked Questions about grill-me

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

FAQPage Schema
How do I stress-test a system design to uncover hidden assumptions?

Design interrogation works by breaking a plan into a decision tree and asking targeted questions for each branch. This relentless questioning forces explicit decisions and surfaces hidden assumptions during architecture planning.

What's the best way to resolve dependencies across interrelated design decisions?

Dependency-driven clarification resolves dependencies between decisions one-by-one to prevent downstream surprises. It applies stepwise question traversal and branch reconciliation to ensure every interrelated choice is explicitly validated.

Can I use codebase exploration to answer design review questions instead of guessing?

Yes, code-first verification switches to exploring the codebase rather than guessing. If a design review question can be answered by reading the repository, the system explores the code directly to verify the assumption.

Does this approach work for clarifying requirements in recursive architecture planning?

Yes, requirements clarification works for recursive architecture planning by applying relentless design interrogation. It drills into every dependency and assumption until both you and the system explicitly agree on every open decision branch.

When should I not use a decision tree for assumption testing?

Avoid using a decision tree for assumption testing when your plan lacks explicit decision branches or interrelated choices. The stepwise question traversal requires a structured plan to break down and reconcile conflicting branches effectively.