One-click install
npx skills add https://github.com/LumenYoung/dotfiles --skill grill-with-docs-lumenyoung
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grill-with-docs
Source: https://github.com/LumenYoung/dotfiles/tree/main/pi-agent/skills/grill-with-docs
Command: npx skills add https://github.com/LumenYoung/dotfiles --skill grill-with-docs-lumenyoung

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It resolves unclear or conflicting project plans by forcing rigorous, terminology-aware questioning and then capturing the resulting decisions directly into the documentation that teams actually rely on.

Core Features & Use Cases

  • Deep plan clarification: Interview-driven resolution that walks the design tree and resolves dependencies between decisions one-by-one.
  • Documentation-aware grilling: Uses existing domain model language in CONTEXT.md and updates documentation (CONTEXT.md and ADRs) inline as decisions crystallise.
  • Session continuation support: Recovers prior grill state and continues using the existing decision/question index rather than restarting.
  • Codebase cross-referencing: When a question can be answered by exploring the codebase, it prioritizes code inspection over speculation.

Quick Start

Ask the AI to run a grilling session to challenge my plan against our existing domain model and update our CONTEXT.md and ADRs inline as we reach confirmed decisions.

Frequently Asked Questions about grill-with-docs

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

FAQPage Schema
How do I stress-test a project plan against an existing domain model?

Stress-testing a plan against a domain model uses interview-driven questioning to walk the design tree and resolve decision dependencies one-by-one. It clarifies terminology and cross-references the codebase to reduce uncertainty before finalizing choices.

How do I update CONTEXT.md and ADRs inline during a design review?

Updating CONTEXT.md and ADRs inline happens automatically during the review when confirmed decisions meet the ADR criteria for trade-offs. The documentation is updated directly as terminology conflicts are resolved and decisions crystallize through the questioning process.

What is the best way to detect glossary conflicts and align terminology across project decisions?

Glossary conflict detection is handled by terminology-aware questioning that checks proposed plans against existing domain model language in CONTEXT.md. It resolves inconsistencies by aligning definitions one question at a time before capturing the agreed-upon terms.

Can I resume a design review session without restarting the decision index?

Resuming a design review session is fully supported through session continuation, which recovers the prior grill state and the existing decision index. This allows you to continue incremental decision recovery without losing your previous progress or context.

Does this plan validation approach work without predefined architecture decision records?

Plan validation works without existing ADRs by focusing on CONTEXT.md domain models and codebase inspection. However, inline ADR updates only occur when the resolved trade-offs specifically meet the established ADR criteria during the grilling process.

Why should I use one-question-at-a-time interaction for architecture tradeoff analysis?

One-question-at-a-time interaction with feedback gating ensures thorough architecture tradeoff analysis by forcing rigorous, sequential resolution of decision dependencies. This prevents speculative answers by prioritizing codebase inspection when questions can be answered by exploring the code.