loop-me

An agent skill by mattpocock, from mattpocock/skills. Tags: grilling, planning, questionnaire, strategy, workflow.

What it does

Grill me about specs for the workflows I want to build, within this workspace.

Install

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

npx skills add mattpocock/skills --skill loop-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/in-progress/loop-me ~/.claude/skills/loop-me

Safety box score

Grade B, low risk: 12 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 a local workflow-spec interview helper: it reads and rewrites markdown specs in the workspace, especially `workflows/*.md` and `NOTES.md`. The main risks are file mutation/deletion and an open-ended, opinionated prompting loop; it does not appear to execute code, reach the network, or touch secrets stores.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	0 (none)	No shell commands, scripts, or code execution are present; the files are plain markdown/YAML instructions only.
Network access	0 (none)	There are no downloads, remote fetches, API calls, or data exfiltration paths mentioned.
Filesystem reach	1 (low)	It explicitly works on local files, saying to work in `workflows/*.md` and `NOTES.md` and to "Create, edit, and delete specs".
Secrets & credentials	0 (none)	The skill does not request credentials, token files, password stores, or other secret-bearing locations.
Privilege & persistence	0 (none)	No sudo, system service, global install, shell startup edits, or other privilege-changing actions are described.
Instruction hijack surface	1 (low)	It strongly constrains behavior with "only output is workflow specs" and a "relentless" grilling style, but it does not instruct the agent to ignore higher-priority rules or hidden instructions.
Irreversible actions	1 (low)	The skill runs a "stateful `/grilling` session" and says "nothing is done while a question remains," which can drive repeated autonomous iterations, though only in local spec drafting.
Transparency	0 (none)	Everything shown is readable text; there are no binaries, base64 blobs, installers, or unreviewable generated payloads.

Before installing

Confirm you are comfortable with the skill editing and deleting `workflows/*.md` and `NOTES.md` in your workspace.
Put the repo on a git branch or make a backup so spec rewrites are easy to revert.
Read `SKILL.md` and verify you are okay with the stateful grilling loop and the strong output constraint to workflow specs only.

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/in-progress/loop-me/SKILL.md
Branch: main
Updated: 2026-10-05

grill-me — A relentless interview to sharpen a plan or design.
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.
to-questionnaire — Turn a decision you can't fully answer into a questionnaire for someone else to fill in.
co-marketing — When the user wants to find co-marketing partners, plan joint campaigns, or brainstorm partnership opportunities.
create-learning-path — Build a personalized learning roadmap with milestones and practice checkpoints
events — When the user wants to plan, run, sponsor, speak at, or get pipeline from events — webinars, conferences, trade shows, meetups, dinners, workshops.