grilling

Interviews the user question-by-question to stress-test plans and designs before implementation.

Updated Jul 3, 2026
One-click install
npx skills add https://github.com/alex-jordan547/agent-setup --skill grilling-alex-jordan547
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grilling
Source: https://github.com/alex-jordan547/agent-setup/tree/main/skills/grilling
Command: npx skills add https://github.com/alex-jordan547/agent-setup --skill grilling-alex-jordan547

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Plans and designs often contain hidden assumptions, unresolved dependencies, and unexplored branches that only surface during implementation. This Skill stress-tests a plan through relentless one-at-a-time questioning before any code is written. ## Core Features & Use Cases - Systematic Interrogation: Walks down each branch of the design tree, resolving dependencies between decisions one by one. - Recommended Answers: Provides a recommended answer with every question so the user can accept, reject, or refine quickly. - Codebase-Grounded Questions: Explores the codebase directly when a question can be answered by reading code instead of asking the user. - Use Case: Before building a new feature, activate this Skill to have every architectural decision, edge case, and dependency challenged in a structured French-language interview until a shared understanding is reached. ## Quick Start Ask the agent to grill you on your implementation plan for the new feature before writing any code.

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 plan before writing code?

Use a structured interrogation approach where each design decision is questioned one at a time, walking down every branch of the design tree. Each question should come with a recommended answer so you can quickly accept or challenge it.

How to validate a software design through questioning?

Resolve dependencies between decisions one by one rather than asking everything at once, which overwhelms the reviewer. Questions answerable from the codebase should be resolved by exploring the code directly instead of asking the user.

When should I use plan interrogation instead of just building?

Use it before implementing any non-trivial feature where architectural choices, edge cases, or dependencies are unclear. It is most valuable when a wrong assumption would be expensive to fix after implementation.

What are the limitations of question-based plan review?

It depends on the user answering each question, so it is slower than autonomous planning and unsuitable for trivial tasks. It also cannot catch issues that neither the questioner nor the user anticipates.