Skills Repo banner
innovationengines innovationengines

Skills Repo

Security community

Description

Five self-contained agent skills for Claude — interview before building, restructure brain dumps, coach beginners, author new skills, and audit your skill library. Install docs for Claude Code, Claude apps, Codex, and ChatGPT.

Installation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open the source below and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

README

skills-repo

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)

Five agent skills for Claude — a small, opinionated set focused on the part of AI work that actually fails: **specification**. Most bad output isn't a model problem, it's a "you asked for 20% of what you meant" problem.

Three skills fix your input before the model acts (`grill-me`, `prompt-master`, `how-to`). Two skills help you build and maintain skills themselves (`write-a-skill`, `skill-audit`).

Each skill is a single self-contained `SKILL.md` — no scripts, no dependencies, no bundled assets. That means they install anywhere, including on platforms that have no formal skill system (see [Install for OpenAI](#install-for-openai)).

Quick install

All five skills into Claude Code, globally:

git clone https://github.com/innovationengines/skills-repo.git /tmp/skills-repo && mkdir -p ~/.claude/skills && cp -r /tmp/skills-repo/*/ ~/.claude/skills/

Windows PowerShell:

git clone https://github.com/innovationengines/skills-repo.git "$env:TEMP\skills-repo"; New-Item -ItemType Directory -Force "$env:USERPROFILE\.claude\skills" | Out-Null; Get-ChildItem "$env:TEMP\skills-repo" -Directory -Exclude .claude | Copy-Item -Destination "$env:USERPROFILE\.claude\skills\" -Recurse -Force

Start a new session, then run `/skills` to confirm. For project-scoped installs, the Claude apps, the Agent SDK, or OpenAI platforms, see [Install for Claude](#install-for-claude) and [Install for OpenAI](#install-for-openai).


The skills

`grill-me` — interview before building

Stops the agent from building on a one-line request. Before producing any code, file, or deliverable, it runs a **10–15 question interview** in 2–3 batched rounds, then confirms a short written build spec, then builds.

  • Triggers on: any non-trivial build/create/make/design/draft request — especially vague ones like "make me a dashboard" or "build an app for my gym."
  • **Won't