name-cop
v1.0.0
Published
Install the Name Cop skill for Claude. Kills your weak name candidates instead of ranking them, applying the SMILE and SCRATCH frameworks created by Alexandra Watkins in Hello, My Name Is Awesome.
Maintainers
Readme
Name Cop
A Claude Code skill that names things properly, by killing your bad candidates instead of ranking them.
The method here is Alexandra Watkins' work, not mine. SMILE and SCRATCH come from her book Hello, My Name Is Awesome (Berrett-Koehler Publishers). This repository applies her frameworks and claims no authorship of them. If they are useful to you, buy her book, which carries the reasoning, the case studies, and the craft that no checklist can. Full attribution and the copyright boundary:
NOTICE.md. Unaffiliated and unendorsed.
Most naming tools hand you a list of two hundred words. This one asks what you are naming and who it is for, generates candidates across several different angles, then runs every one through seven deal breakers. Anything that trips one is dead on the spot. Only the survivors get scored.
This is her method, encoded so an agent will actually hold you to it.
Install
In Claude Code, as a plugin. This is the one to use: it brings the skill and the
/name-cop-init command.
/plugin marketplace add jainulabudeenm/name-cop
/plugin install name-copOr from npm, if you want the skill in Claude.ai or anywhere outside Claude Code. This installs the skill on its own, without the slash command.
npx name-copAdd --project to install into the current repo instead of your home directory,
so you can commit it and share it with a team.
Use
Say what needs a name.
> I need a name for a tool that turns receipt photos into a tax-ready spreadsheet.On first run it asks six short questions and saves a brief to ~/.claude/name-cop/profile.md: what
the thing is, who it is for, what it must not sound like, what is already taken, and your hard
constraints. Everything after that starts from the brief.
Re-run /name-cop-init when you are naming something new.
Why a kill rule
The failure mode in naming is not a shortage of ideas. It is falling for your own first idea and then defending it. So the process is deliberately lopsided:
SCRATCH, the seven deal breakers, run first. Spelling-challenged, Copycat, Restrictive, Annoying, Tame, Curse of knowledge, Hard to pronounce. Trip any one and the candidate is out. Not weakened, not conditional, not rescued by a good logo.
SMILE, the five marks, run second, on survivors only. Suggestive, Memorable, Imaginative, Legs, Emotional. Scored 0 to 2 each, out of ten, stated openly.
That order matters. Score first and a high total will always find a way to argue a fatal flaw down.
The kill that gets argued with most is Tame, because a flat name feels like the responsible choice. It is not. A name that describes the category is a label, and the skill will keep saying so.
You get three ranked finalists, each with its score and its weakest mark named out loud. When you pick one, it makes the case against your pick once, then logs it and stops.
What it will not do
- Check availability. No live domain or registry access. It tells you exactly what to check, the exact-match domain, the package name, the handle, the trademark class, and never claims a name is free.
- Trademark clearance. Not legal advice. For a real launch that belongs with a lawyer.
- Fix an unclear product. If the brief shows the thing itself cannot be described in one sentence, it will say so. No name saves that.
The brief
~/.claude/name-cop/profile.md, in your home directory rather than a project folder, so an unshipped
name does not get committed to a public repo by accident. Plain markdown, edit it whenever. See
profile.example.md for the level of detail that produces useful candidates.
It also keeps the killed list, so a name you rejected last week does not come back this week in a different font.
Credit, and one contributing rule
SMILE and SCRATCH are Alexandra Watkins' frameworks, published in her book Hello, My Name Is Awesome (Berrett-Koehler Publishers). She created this method. This repository applies it and takes no credit for it. Buy the book.
It encodes the method only, in its own words. It does not carry the book's text, its example name lists, its exercises, or its brainstorming word banks, and pull requests that add any of that will be closed. Every example here is written fresh. If you want to contribute a better explanation of a SCRATCH letter, write it yourself.
Full attribution statement: NOTICE.md. This is an unaffiliated personal project, not
sponsored by or endorsed by Alexandra Watkins or her firm.
License
MIT
