grill-me

An agent skill by mattpocock, from mattpocock/skills. Tags: brainstorming, grilling, review, strategy, workflow.

What it does

A relentless interview to sharpen a plan or design.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add mattpocock/skills --skill grill-me

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/mattpocock/skills
cp -r skills/skills/productivity/grill-me ~/.claude/skills/grill-me

Safety box score

Grade A, minimal risk: 0 of 100 risk points, rated 2026-10-16. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill is just a conversational coaching prompt: it asks the agent to run a "grilling" interview and contains no code, scripts, downloads, or file access. The only meaningful risk is behavioral—being relentlessly interrogative—but it does not expand access to the machine, accounts, or data.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	0 (none)	No scripts or executable code are present; `SKILL.md` only says `Call the Skill tool with "grilling".`
Network access	0 (none)	There are no URLs, download commands, API calls, or other remote data flows in the two text files.
Filesystem reach	0 (none)	The skill does not read, write, or delete files; it only defines a coaching prompt in `SKILL.md` and `agents/openai.yaml`.
Secrets & credentials	0 (none)	It does not request tokens, passwords, `.env` files, keychains, or browser sessions.
Privilege & persistence	0 (none)	There is no sudo, installation, service registration, shell startup editing, or other privilege-changing behavior.
Instruction hijack surface	0 (none)	The files do not ask the agent to ignore its rules, hide actions, or follow external instructions; `disable-model-invocation: true` and `allow_implicit_invocation: false` actually constrain behavior.
Irreversible actions	0 (none)	It is an interview skill only; it does not post, deploy, delete, pay, or otherwise act on the user's behalf.
Transparency	0 (none)	Everything is plain text and fully readable; there are no binaries, installers, or decoded-at-runtime payloads.

Before installing

Verify the repo only contains the two plain-text files shown here (`SKILL.md` and `agents/openai.yaml`).
Confirm you want a purely conversational skill that will challenge your plan, not automate any actions.
No special system or account checks are needed before using this skill.

Rated 2026-10-16 by an automated review of 2 files from the skill's repository.

Source

Repository: mattpocock/skills (all skills from this repository)
Path: skills/productivity/grill-me/SKILL.md
Branch: main
Updated: 2026-10-05

grilling — 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.
principle-attack-the-premise — Apply when two or more fixes that share one premise have failed the same gate.
grill-with-docs — A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.
loop-me — Grill me about specs for the workflows I want to build, within this workspace.
advisor — Advisor mode. Consult a stronger (or different) model at key checkpoints: before major decisions, when stuck on an error, and before declaring a task done.
goal-prompt — Drafts copy-paste-ready /goal commands for goal mode in Claude Code and Codex.