grill-me

Surface unresolved decisions and gaps in feature specs through adversarial Q&A.

181|29|Updated Apr 21, 2026

Search Tags:#documentation #qa #spec #pm #decision-analysis #grooming #grill

Ready to run

npx skills add https://github.com/iusztinpaul/squid --skill grill-me-iusztinpaul

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: grill-me
Source: https://github.com/iusztinpaul/squid/tree/main/skills/grill-me
Command: npx skills add https://github.com/iusztinpaul/squid --skill grill-me-iusztinpaul

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Grill-me helps product teams prevent misaligned feature specs by conducting adversarial Q&A that surfaces unresolved decisions, risky assumptions, and gaps before PM grooming.

Core Features & Use Cases

Stepwise grilling workflow that identifies modal terms, edge cases, and decision points in a spec.
Generates a grilled, decision-rich spec with explicit decisions and open questions for PM review.
Integrates a reproducible process: resolve the spec, reconnoitre the codebase, build a question matrix, draft answers, and synthesize the output.
Produces a shareable artifact ready for PM grooming, reducing back-and-forth and surfacing risks early.

Quick Start

Provide a feature spec (or path to a markdown spec) and run /grill-me to generate a grilled, decision-rich spec for PM grooming.

Frequently Asked Questions about grill-me

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

FAQPage Schema

How do I surface unresolved decisions in a feature spec before PM grooming?▼

Adversarial Q&A surfaces unresolved decisions in a feature spec by identifying modal terms, edge cases, and risky assumptions before PM grooming. This stepwise grilling workflow produces an explicit, decision-rich grilled spec with traceable rationale.

What is the best way to prepare a feature spec for product management review?▼

Preparing a feature spec for product management review involves resolving the spec, reconnoitering the codebase, building a question matrix, and drafting concrete decisions. This generates a shareable, grilled artifact with explicit open questions for PM review.

How does adversarial Q&A prevent misaligned feature specs?▼

Adversarial Q&A prevents misaligned feature specs by conducting stepwise questioning that extracts decision points and gaps early. This reduces back-and-forth during development by delivering an explicit, decision-rich spec ready for PM grooming.

Can I use a markdown file path to start the spec grilling process?▼

Yes, you can provide a feature spec or a path to a markdown spec to start the grilling process. The workflow resolves the provided spec, reconnoitres the codebase, and builds a question matrix to draft concrete decisions for PM grooming.

What does a grilled spec output contain for decision analysis?▼

A grilled spec output contains explicit decisions, traceable rationale, and open questions for decision analysis. It synthesizes answers from a built question matrix into a shareable artifact designed specifically for PM grooming.