grill-me

Interrogate plans and designs to expose missing assumptions and weak decision points.

1|Updated Jun 27, 2026

Search Tags:#planning #architecture #interview #codebase analysis #assumptions #design review

Ready to run

npx skills add https://github.com/babually/bongo-template --skill grill-me-babually

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: grill-me
Source: https://github.com/babually/bongo-template/tree/main/.agents/skills/grill-me
Command: npx skills add https://github.com/babually/bongo-template --skill grill-me-babually

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you pressure-test a plan, design, or proposal by systematically probing assumptions, dependencies, edge cases, and tradeoffs until the approach is clear and defensible.

Core Features & Use Cases

Relentless Interviewing: Asks one question at a time to uncover missing details and resolve ambiguity.
Decision-Tree Exploration: Walks through branches of a design or plan in dependency order so decisions are made coherently.
Codebase-Aware Follow-Up: Checks the repository when a question can be answered from existing code rather than forcing speculation.
Use Case: Use it when you want an AI to challenge your architecture, product plan, or implementation strategy before you commit to it.

Quick Start

Ask the assistant to grill your plan or design and keep questioning you until the approach is fully understood.

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 an architecture plan to expose missing assumptions?▼

Architecture review through structured adversarial questioning exposes missing assumptions and unresolved dependencies by asking one question at a time. It walks through decision branches sequentially, checking the codebase directly when evidence is needed, ensuring your design holds up under pressure.

What is the best way to interrogate a product plan for unresolved dependencies?▼

Interrogating a product plan requires relentless one-question-at-a-time interviewing that explores each decision-tree branch in dependency order. This structured adversarial approach uncovers unresolved dependencies and weak decision points, forcing coherent resolution of tradeoffs before commitment.

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

Yes, design review questions that can be answered from existing code are resolved through direct codebase exploration. The Skill checks the repository for evidence rather than forcing speculation, grounding architecture decisions in actual implementation state.

When do I need structured adversarial questioning for an implementation strategy?▼

You need structured adversarial questioning for an implementation strategy when you want to pressure-test decisions before committing. It applies to any discussion benefiting from branch-by-branch resolution of tradeoffs, edge cases, and missing assumptions in your plan or design.

What are the limitations of using one-question-at-a-time interviewing for design reviews?▼

The limitation of one-question-at-a-time interviewing is that it requires branch-by-branch resolution, which can be time-intensive for complex architecture reviews. This thoroughness is necessary to ensure no weak decision points or unresolved dependencies survive the interrogation.