One-click install
npx skills add https://github.com/luan/agents --skill grill-luan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grill
Source: https://github.com/luan/agents/tree/main/skills/grill
Command: npx skills add https://github.com/luan/agents --skill grill-luan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

It prevents misalignment between a proposed plan and your project’s established domain language, terminology, and recorded decisions by forcing clarification and validation against the codebase and docs.

Core Features & Use Cases

  • Relentless plan interrogation: Walks each branch of the design, asking one question at a time and waiting for feedback before proceeding.
  • Glossary and context conflict detection: Immediately challenges ambiguous or conflicting terms by comparing user wording against CONTEXT.md.
  • Live updates to domain docs: Resolves terms and updates CONTEXT.md inline using the project’s required format, and offers ADRs only when a decision meets the stated trade-off criteria.

Quick Start

Ask: "Use grill to interview and stress-test this plan against our existing domain model and recorded decisions, question-by-question, updating our context and ADRs when justified."

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 proposed plan against existing domain models and decision records?

To stress-test a plan against domain models and decision records, you need an interrogation workflow that cross-checks proposed terminology and dependencies against your codebase and docs, asking one question at a time to uncover ambiguity and validate architecture trade-offs.

What is the best way to detect glossary conflicts during design walkthroughs?

The best way to detect glossary conflicts during design walkthroughs is by comparing user wording against an established CONTEXT.md file, immediately challenging ambiguous terms to ensure domain terminology aligns with recorded decisions and codebase reality.

How do I validate architecture trade-offs and update decision records inline?

You validate architecture trade-offs by interrogating plan branches against existing decision artifacts, then updating ADRs inline only when a decision meets stated criteria, while resolving terms directly within CONTEXT.md using required formats.

Can I use iterative interrogation to clarify requirements without overwhelming the design process?

Yes, you can clarify requirements without overwhelming the process by applying iterative, single-question-at-a-time interrogation, waiting for feedback before proceeding to the next branch of the design walkthrough.

When do I need to update CONTEXT.md and ADRs during domain modeling?

You need to update CONTEXT.md and ADRs during domain modeling when a stress-test uncovers misalignment or a decision meets stated trade-off criteria, resolving terms inline to prevent ambiguity between the proposed plan and established project docs.