Interviews the user question-by-question to stress-test plans and resolve design decisions.
npx skills add https://github.com/lrstanley/skills --skill grill-me-lrstanley
Please help me install this Agent Skill. Skill: grill-me Source: https://github.com/lrstanley/skills/tree/main/grill-me Command: npx skills add https://github.com/lrstanley/skills --skill grill-me-lrstanley
(preferred) prefix and placed first when using structured question tools. - Codebase-Aware Questioning: Explores the codebase directly to answer questions that can be resolved from existing code instead of asking the user. - Use Case: Before starting a major refactor, ask to be grilled on your migration plan so every architectural decision, dependency, and edge case is explicitly resolved before writing code. ## Quick Start Ask the assistant to grill you on your plan or design document and answer each question as it comes.High-intent search queries and answers about installing and using this skill.
Ask to be grilled on your plan and the assistant will interview you one question at a time, walking each branch of the decision tree. Each question includes a recommended answer so you can accept or challenge it until every decision is resolved.
It asks questions one at a time rather than in batches. When structured question tools with pre-configured options are available, the recommended answer is placed first and prefixed with (preferred) in the option label.
No. If a question can be answered by exploring the codebase, the assistant explores the code directly instead of asking. Questions are reserved for decisions only the user can resolve.
Use plan grilling before implementation to resolve design decisions and hidden assumptions interactively. Standard code review happens after code exists and focuses on correctness rather than decision-tree exploration.
It depends on the user's answers and cannot verify feasibility against external systems or runtime behavior. It resolves decision logic and shared understanding, not empirical performance or integration testing.
quaid-appDraft structured architectural proposals with problem statements, decisions, and risks.
OmniNode-aiCreate Linear tickets from markdown plan files with dependencies and epics.
abagamesAnalyze the preceding response for hidden assumptions, logical gaps, and AI-specific failure modes.
betaTridentCapture and maintain structured architectural decision records using a standard ADR template.