grill

Stress-tests plans and designs through one-question-at-a-time interrogation before coding begins.

Updated Dec 1, 2025
One-click install
npx skills add https://github.com/manusco/wolfgang --skill grill-manusco
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: grill
Source: https://github.com/manusco/wolfgang/tree/main/.cursor/skills/grill
Command: npx skills add https://github.com/manusco/wolfgang --skill grill-manusco

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Plans and designs often contain hidden assumptions, gaps, and contradictions that only surface during implementation, causing costly rework. This Skill stress-tests a plan or design through relentless one-question-at-a-time interrogation before any code is written, building shared understanding early. ## Core Features & Use Cases - Pre-Code Stress Testing: Interrogates a plan or design to expose weak assumptions and unresolved decisions before implementation starts. - One-Question-at-a-Time Method: Asks focused sequential questions rather than overwhelming reviews, driving toward shared understanding. - Definition of Done: Follows a structured procedure with explicit completion criteria routed from the /grill command. - Use Case: Before starting a new feature, run /grill on your design document so the AI challenges each decision point until the plan is airtight. ## Quick Start Ask the AI to run the grill procedure on your current plan or design document to stress-test it before writing any code.

Frequently Asked Questions about grill

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

FAQPage Schema
How do I stress-test a design plan before coding?▼

Run the grill procedure on your plan or design document. It interrogates the plan one question at a time, exposing weak assumptions and unresolved decisions until shared understanding is reached, following its Definition of Done.

What is the one-question-at-a-time interrogation method?▼

It is a focused review technique where the AI asks a single targeted question at a time rather than delivering a bulk critique. This keeps each decision point explicit and drives the discussion toward concrete resolution.

When should I use plan interrogation instead of code review?▼

Use it before any code exists, when you want to catch design flaws cheaply. Code review catches implementation errors; plan interrogation catches flawed assumptions and missing decisions upstream.

Does the grill skill write or modify code?▼

No. It operates strictly before coding begins, routing the /grill command to a procedure file that interrogates plans and designs. Its output is shared understanding and a validated plan, not code changes.

What are the limitations of interrogation-based plan review?▼

It depends on the quality of the plan provided and the answers given; vague inputs yield shallow interrogation. It also does not replace prototyping or testing for validating technical feasibility.