personal-grill-me

Interviews plans, designs, and ideas in Japanese through round-based frontier questioning.

Updated Apr 25, 2026
One-click install
npx skills add https://github.com/kosako/agent-tools --skill personal-grill-me-kosako
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: personal-grill-me
Source: https://github.com/kosako/agent-tools/tree/main/shared/skills/personal-grill-me
Command: npx skills add https://github.com/kosako/agent-tools --skill personal-grill-me-kosako

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Plans, designs, and product ideas often contain vague assumptions, contradictions, and untested hypotheses that surface too late. This Skill relentlessly interviews you in Japanese, round by round, until every implicit assumption is resolved and a shared understanding is reached. ## Core Features & Use Cases - Round-based frontier questioning: Groups all currently answerable questions into a single numbered round with recommendations, deferring dependent questions to later rounds. - Rigorous challenge of vagueness: Refuses to accept answers like "somehow" or "roughly", challenges assumptions, contradictions, and wishful thinking, and never starts implementation before agreement. - Fact vs. decision separation: Investigates codebase facts and reference materials itself, asking you only for decisions that require judgment. - Use Case: You want to build a new authentication system with JWT. The Skill batches questions about session management, refresh tokens, logout, and security edge cases into one round, each with a recommended answer, and pushes back on any vague replies. ## Quick Start Ask the AI to grill you on your project plan or design idea using the personal-grill-me skill until you reach a shared understanding.

Frequently Asked Questions about personal-grill-me

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

FAQPage Schema
How do I stress-test a product idea or design with AI?▼

Use an interview-style skill that questions your plan in rounds, attaching its own recommendations to each question. It challenges vague answers, contradictions, and untested assumptions until every implicit decision is made explicit and agreed upon.

What is round-based frontier questioning in AI interviews?▼

Frontier questioning batches all questions whose prerequisites are already settled into one numbered round, each with a recommended answer. Questions depending on unanswered ones are deferred to later rounds, so the conversation stays efficient and deep.

Can this skill write documents or ADRs from the interview?▼

No, this skill is read-only and conversation-only; it writes nothing to the repository. If you want glossaries or ADRs produced alongside the interview, use the companion skill personal-grill-with-docs instead.

When should I not use an interview-style grilling skill?▼

Avoid it for simple factual questions, direct implementation requests, or document creation tasks. It is designed for explicit brainstorming, plan validation, and decision-making work where the goal is reaching shared understanding, not producing artifacts.

Does the AI ask about facts it could look up itself?▼

No, the skill investigates codebase structure, existing implementations, libraries, and reference materials on its own. It only asks you for decisions requiring judgment, and it keeps questioning without blocking on ongoing fact-finding.