batch-grill-me

Interviews users in rounds of frontier questions to refine a design plan.

Updated Jul 30, 2026
One-click install
npx skills add https://github.com/j172/bid --skill batch-grill-me-j172
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: batch-grill-me
Source: https://github.com/j172/bid/tree/main/agent/skills/batch-grill-me
Command: npx skills add https://github.com/j172/bid --skill batch-grill-me-j172

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Plans and design decisions often contain hidden assumptions and unresolved dependencies. This Skill structures a relentless interview that maps every decision as a design tree and asks all currently answerable questions in batched rounds, so nothing is left silently assumed before acting. ## Core Features & Use Cases - Design Tree Mapping: Models every decision as a branch, tracking which questions depend on which answers. - Frontier Batching: Asks every question whose prerequisites are settled in a single numbered round, each with a recommended answer. - Fact Delegation: Dispatches sub-agents to look up environmental facts instead of asking the user for anything discoverable. - Use Case: Before implementing a new feature, run the interview to surface every open design decision round by round until the frontier is empty and a shared understanding is confirmed. ## Quick Start Ask the assistant to grill you in rounds of questions about your plan until every design decision is settled.

Frequently Asked Questions about batch-grill-me

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

FAQPage Schema
How do I refine a plan through structured AI questions?

Run a round-based interview where the assistant maps your decisions as a tree and asks every currently answerable question in one numbered batch. Each of your answers settles decisions and unblocks the next round of dependent questions.

What is a frontier question in design planning?

A frontier question is a decision whose prerequisites are already settled, meaning it can be asked now without guessing at unanswered questions. Questions depending on still-open decisions are deferred to later rounds.

Can the interview look up facts instead of asking me?

Yes. When a question needs a fact from the filesystem or tools, a sub-agent is dispatched to find it. Only decisions are put to the user; discoverable facts are never asked.

When does the question-and-answer session end?

The session ends when the frontier is empty, meaning every branch of the design tree has been visited and nothing remains silently assumed. The assistant waits for your confirmation of shared understanding before acting.

When should I not use round-based interviewing?

Avoid it for trivial tasks with no branching decisions or when requirements are already fully specified. The method is designed for plans with interdependent open decisions, not single-step execution.