grill-me

Interrogates underspecified ideas through focused, one-question-at-a-time discovery interviews.

1|Updated Mar 6, 2026
One-click install
npx skills add https://github.com/grant-vine/wunderkind --skill grill-me-grant-vine
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grill-me
Source: https://github.com/grant-vine/wunderkind/tree/main/skills/grill-me
Command: npx skills add https://github.com/grant-vine/wunderkind --skill grill-me-grant-vine

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps teams quickly uncover gaps and ambiguities in ideas by guiding a relentless interviewer who asks precise, focused questions until the problem is clearly defined.

Core Features & Use Cases

  • Ask one sharp question at a time to collapse ambiguity and surface hidden decisions.
  • Prefer questions that reveal tradeoffs and boundaries of the problem.
  • If the user goal can be learned from the repo, inspect the repo instead of asking.
  • Continue questioning until tradeoffs, scope, and success criteria are explicit.
  • Summarize the evolving problem statement every few turns.

Quick Start

Provide a brief problem statement and I will begin with a sharp clarifying question.

Frequently Asked Questions about grill-me

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

FAQPage Schema
How do I clarify ambiguous product requirements before development?

Product scope discovery works best when questions are asked one at a time, revealing tradeoffs and boundaries until the problem statement, scope, and success criteria are fully explicit.

What's the best way to run a discovery interview for underspecified features?

Run a discovery interview by interrogating underspecified features with one precise question at a time, summarizing the evolving problem statement every few turns to collapse ambiguity and record decisions.

Can I use this to define success criteria and scope for architecture directions?

Yes, this can define success criteria and scope for architecture directions by relentlessly questioning underspecified plans until tradeoffs, boundaries, and success criteria are fully explicit and recorded.

How do I uncover hidden tradeoffs and decisions during product discovery?

Uncover hidden tradeoffs and decisions during product discovery by preferring interview questions that reveal problem boundaries, enforcing a one-question-at-a-time approach to prevent oversight.

Does requirements discovery work if my user goal can be learned from the code repository?

If the user goal can be learned from the repository, the discovery process will inspect the repository directly instead of asking unnecessary questions to define the product requirements.

When should I not use a one-question-at-a-time approach for product clarification?

You should avoid a one-question-at-a-time approach for product clarification if your idea is already fully specified with explicit tradeoffs, scope, and success criteria, as further interviewing would be redundant.