grill-me

Resolve under-specified design and migration plans through sequential decision interrogation.

1|Updated May 8, 2026
One-click install
npx skills add https://github.com/BioInfo/rundatarun --skill grill-me-bioinfo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grill-me
Source: https://github.com/BioInfo/rundatarun/tree/main/skills/grill-me
Command: npx skills add https://github.com/BioInfo/rundatarun --skill grill-me-bioinfo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents expensive build failures caused by committing to an under-specified plan by forcing every open design decision to be clarified before you move forward.

Core Features & Use Cases

  • One-question-at-a-time grilling: Interrogates your plan branch-by-branch and waits for your answer before continuing.
  • Recommended answer per decision: Provides a concrete suggested response with a brief reason so you react to positions, not blanks.
  • Decision-tree resolution with stop condition: Continues until all branches are resolved and shared understanding is established.
  • Reads code when discoverable: Uses the codebase as the source of truth when the answer can be found there.
  • Closing artifact: Produces a short bullet list of Resolved decisions plus an Open/deferred list.

Quick Start

Trigger the skill by telling your AI to run "grill me on this migration plan" and then answer each question it asks until it outputs the Resolved decisions summary.

Frequently Asked Questions about grill-me

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

FAQPage Schema
How do I resolve under-specified software migration plans before implementation?

Resolve under-specified software migration plans by interrogating open design decisions one at a time. The grilling process waits for your answer at each branch and provides recommended responses until every decision is settled, preventing expensive build failures.

What is the best way to review a software design decision tree for a new service?

Reviewing a software design decision tree is best handled by sequentially interrogating each open branch. The process asks one question at a time, reads the codebase for discoverable answers, and outputs a final summary of resolved and deferred decisions.

Can I use codebase reasoning to clarify open decisions during pre-implementation planning?

Yes, you can use codebase reasoning to clarify open decisions during pre-implementation planning. When answers are discoverable in the code, the codebase is read directly as the source of truth to resolve design branches without relying solely on manual input.

How do I force design decisions for a non-trivial schema change before coding starts?

Force design decisions for a non-trivial schema change by triggering a sequential question and answer flow. The interrogation continues branch-by-branch, offering recommended answers with brief reasons, until all open decisions are resolved and a plan solidifies.

When do I need decision tree interrogation for a refactor plan?

You need decision tree interrogation for a refactor plan when the design is under-specified and you must force open decisions before moving forward. It is specifically for non-trivial refactors where unresolved branches could cause expensive build failures.

What output artifact do I get after interrogating all open design decisions?

After interrogating all open design decisions, you get a closing artifact containing a short bullet list. This summary includes all resolved decisions alongside an open or deferred list, establishing shared understanding before implementation begins.