grilling

Interviews the user in structured rounds to resolve every branch of a design decision tree.

Updated Apr 6, 2026
One-click install
npx skills add https://github.com/toderian/project_template --skill grilling-toderian
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grilling
Source: https://github.com/toderian/project_template/tree/main/plugins/agents-core/skills/grilling
Command: npx skills add https://github.com/toderian/project_template --skill grilling-toderian

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Plans and design decisions often hide unexamined assumptions that surface late as rework. This Skill stress-tests an idea by systematically questioning every open decision until nothing is left silently assumed. ## Core Features & Use Cases - Design tree mapping: Models the topic as a tree of decisions where each answer unlocks the dependent questions behind it. - Frontier-based rounds: Asks all currently answerable questions in one numbered round, each with a recommended answer, then recomputes the frontier after the user replies. - Fact lookup delegation: Dispatches sub-agents or direct lookups for environmental facts (code, tests, docs, web) so the user is only asked for decisions, never research. - Use Case: Before implementing a new authentication system, run a grilling session to settle questions about session storage, token expiry, and provider choice, ending with a summary of decisions made, open questions, and risks. ## Quick Start Ask the agent to grill you on your plan for the upcoming feature until every design decision is resolved.

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 or design decision with an AI agent?

Invoke the grilling skill and state the plan or idea you want examined. The agent maps it as a decision tree and asks numbered questions in rounds, each with a recommended answer, until every branch is resolved.

How does the grilling skill decide which questions to ask first?

It computes a frontier: the set of decisions whose prerequisites are already settled. Questions depending on unanswered questions in the current round are deferred to later rounds, so nothing is asked prematurely.

Does the grilling skill ask the user for facts about the codebase?

No. Finding facts is the agent's job, not the user's. When a question needs information from code, tests, docs, or the web, the agent dispatches a research sub-agent or looks it up itself and only asks the user for decisions.

When does a grilling session end?

The session ends when the frontier is empty, meaning every branch of the design tree has been visited. The agent closes with a summary of decisions made, open questions deferred on purpose, and risks identified, then waits for user confirmation.

When should I not use a grilling-style interview?

Avoid it for trivial or fully specified tasks where no open decisions exist, since the multi-round interview adds overhead. It is designed for ambiguous plans, architecture choices, and ideas that need assumption surfacing.