Skip to content
Grilling

Grilling Skill

Matt Pocock's grilling skill: the agent interviews you relentlessly about a plan or decision, mapping it as a design tree and asking every answerable question per round, with a recommended answer for each.

GitHub Docs

Install

Install with the skills CLI

One command for Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf and dozens of other agents. Needs Node.js; the skill is fetched straight from mattpocock/skills.

$ npx skills add mattpocock/skills --skill grilling

Run it in a terminal. The CLI finds the agents installed on your machine, asks which ones to add the skill to, and whether to install it into the folder you ran it from or for all your projects.

Other ways to install

Copy the whole skill folder — SKILL.md and any scripts or references beside it — into ~/.claude/skills/grilling/, where Claude Code picks it up in every project. Only git is needed.

tmp=$(mktemp -d) && \
  git clone --depth 1 --branch main https://github.com/mattpocock/skills.git "$tmp" && \
  mkdir -p ~/.claude/skills && \
  cp -R "$tmp"/skills/productivity/grilling ~/.claude/skills/grilling

No scripts or references next to it? Saving the SKILL.md alone as ~/.claude/skills/grilling/SKILL.md works too.

Using another agent? Its personal skills folder
Codex
~/.agents/skills/
Cursor
~/.cursor/skills/
GitHub Copilot
~/.copilot/skills/
Gemini CLI
~/.gemini/skills/
Windsurf
~/.codeium/windsurf/skills/
OpenCode
~/.config/opencode/skills/

About

  • Maps the plan as a design tree of dependent decisions
  • Asks the whole answerable frontier each round as numbered questions
  • Gives a recommended answer per question, worded so yes accepts it
  • Looks up facts itself via sub-agents instead of asking you
  • Ends only when every branch is settled and you confirm

Grilling is the interview engine behind Matt Pocock's /grill-me and /grill-with-docs commands, published on its own so an agent can pick it up whenever someone wants a plan, decision or idea stress-tested. Unlike the two wrapper commands, it can be triggered by the model, not only by the user.

The skill treats the plan as a design tree in which every decision branches into the decisions that depend on it. It works in rounds: the "frontier" is every decision whose prerequisites are already settled, and the agent asks the whole frontier at once as numbered questions, each with its own recommended answer worded so that "yes" accepts it. After each round of answers it recomputes the frontier, so questions that depend on something still open wait for a later round rather than forcing guesses.

A clear division of labour keeps the interview short. Facts are the agent's job: when a question needs something from the codebase or environment, the agent sends a sub-agent with a fresh context to look it up instead of asking you, and keeps asking the questions that do not depend on that lookup. Decisions are yours: each one is put to you and the agent waits. The session ends only when the frontier is empty, with no branch left silently assumed, and the agent does not act on the result until you confirm you share an understanding.

It suits engineers and product people who want to find the holes in a design before code is written, and it pairs with the domain-modeling skill from the same repository. MIT licensed.

SKILL.md

The instructions your agent loads, as published by Matt Pocock.

SKILL.md
name
grilling
description
Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

Interview the user relentlessly until you reach a shared understanding. Map this as a design tree: every decision branches into the decisions that hang off it.

Work the tree in rounds. The frontier is every decision whose prerequisites are already settled: the questions you can ask _now_ without guessing at answers you haven't heard yet. Ask the whole frontier in one round: number each question and give your recommended answer. Then wait for the user's answers before the next round.

Format a round like so:

❓ **Q1** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices>

➡️ <your recommended answer>

---

❓ **Q2** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices>

➡️ <your recommended answer>

Word each question so "yes" accepts your recommended answer.

Each round the user answers reshapes the tree: settled decisions push the frontier outward and unblock questions that depended on them. Recompute the frontier and ask the next round. A question whose answer depends on another question still open in this round belongs to a _later_ round, not this one.

Finding _facts_ is your job, never the user's. When a frontier question needs a fact from the environment (filesystem, tools, etc.), dispatch a sub-agent with a fresh context window to find it; don't ask the user for anything you could look up yourself. Don't block on it: a running exploration is an unsettled prerequisite, so only the questions downstream of it wait for the sub-agent to report; ask the rest of the frontier now. The _decisions_ are the user's: put each to them and wait.

The session is done when the frontier is empty: every branch of the design tree visited, nothing left silently assumed. Do not act on it until the user confirms you have reached a shared understanding.

Install

Skill: Grill Me

by Matt Pocock

New

Matt Pocock's /grill-me command: a relentless interview that stress-tests a plan or design, asking numbered questions with recommended answers in rounds until every decision is settled. It hands off to his grilling skill.

Skill: Brainstorming

by Jesse Vincent

Featured

Explore intent, requirements, and design with the user before any code gets written — the antidote to building the wrong thing quickly.

Planning skill that interrogates your use case before any code is written, then recommends the right Twilio Conversations stack — ConversationRelay, Memory, Intelligence, Orchestrator, TaskRouter — and the implementation skills to follow.

71 views
Browse all skills →