Skip to content
Skills Library
Management & product · Planning

prototype

/prototype

Test an idea in the cheapest way before building it — pick the right level (concept test, fake-door button, clickable screen, bare playable mechanic, small slice in the real game), set a success criterion first, build throwaway prototypes in a separate prototypes/ folder, and record the result as an experiment. Use for "프로토타입", "이 아이디어 먼저 시험해 보고 싶어", "만들기 전에 반응 보고 싶어", "fake door", "quick prototype", "would players want this".

Repository
hype-me ↗
Licence
MIT
Last repository change
13.11.2026

Assessment and security

95/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content30 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 13.11.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Test an idea in the cheapest way before building it — pick the right level (concept test, fake-door button, clickable screen, bare playable mechanic, small slice in the real game), set a success criterion first, build throwaway prototypes in a separate prototypes/ folder, and record the result as an experiment. Use for "프로토타입", "이 아이디어 먼저 시험해 보고 싶어", "만들기 전에 반응 보고 싶어", "fake door", "quick prototype", "would players want this".

Triggers from the description

프로토타입이 아이디어 먼저 시험해 보고 싶어만들기 전에 반응 보고 싶어fake doorquick prototypewould players want this

Arguments

/prototype [idea]
Original author description

Test an idea in the cheapest way before building it — pick the right level (concept test, fake-door button, clickable screen, bare playable mechanic, small slice in the real game), set a success criterion first, build throwaway prototypes in a separate prototypes/ folder, and record the result as an experiment. Use for "프로토타입", "이 아이디어 먼저 시험해 보고 싶어", "만들기 전에 반응 보고 싶어", "fake door", "quick prototype", "would players want this".

How it works

  1. 1. Pin down the question
  2. 2. Pick the cheapest level that answers it
  3. 3. Build (levels 3–4)
  4. 4. Test and decide
Original SKILL.md headings · 29 lines, 490 words
SKILL.md preview
---
name: prototype
description: Test an idea in the cheapest way before building it — pick the right level (concept test, fake-door button, clickable screen, bare playable mechanic, small slice in the real game), set a success criterion first, build throwaway prototypes in a separate prototypes/ folder, and record the result as an experiment. Use for "프로토타입", "이 아이디어 먼저 시험해 보고 싶어", "만들기 전에 반응 보고 싶어", "fake door", "quick prototype", "would players want this".
argument-hint: "[idea]"
---

# Prototype — cheapest test first

Answer in the user's language. Prototypes are throwaway and live apart from the game.
Follow the shared rules in `${CLAUDE_PLUGIN_ROOT}/references/rules.md` — profile first (run setup, then continue), end with one experiment suggestion, use design checks whenever something visual is involved, and leave a runnable test with anything you make.

## 1. Pin down the question
If the idea is still vague, do not stop with only questions: give a first answer in at most 5 lines — the recommended level and why, an example success criterion, that any prototype lives in `prototypes/` apart from the game, and that it will be recorded as an experiment — then ask at most 2 questions (what the idea is, and want / understand / enjoy).
From the idea or an `experiment` record: what do we need to learn — do people **want** it, **understand** it, or **enjoy** it? Set the success criterion before building (e.g. "3 of 5 testers finish 2 rounds", "5%+ click the fake-door button").

## 2. Pick the cheapest level that answers it
| level | answers | how | time |
|---|---|---|---|
| 1 concept test | want? | 2–3 hook lines or thumbnail drafts posted by the user to a small community, or a Discord poll | ~1 h |
| 2 fake door | want enough to click? | a "coming soon" button/page; measure clicks with a tracking link or an analytics event. Clicking must say honestly that it is not ready yet | ~half a day |
| 3 screen prototype | understand? | one clickable HTML screen or flow | half a day–1 day |
| 4 play prototype | enjoy? | a bare playable mechanic in one HTML file | 1–2 days |
| 5 slice in the game | works for real players? | behind a switch for some players | days |
Say why this level and not a cheaper one.

## 3. Build (levels 3–4)
Start the `prototyper` agent with the question, criterion, kind, project path and style notes. It writes only to `prototypes/<name>/`. Then let one `ux-playtester` try it as a first look (from the `ux` skill), if the level allows.
For level 1, write the drafts (the user posts them). For level 2, write the button/page text and the tracking link (a `utm_` link as in `promo-plan`) or event line; changes to the real game happen only after a yes, and never collect payment or personal data behind a fake door.

## 4. Test and decide
Give the user: the link/file, who to show it to, the 3 questions to ask testers, and what to count. Record it as an experiment (`experiment new`, type "prototype test") with the check date. On review: **build / change / drop**, with the numbers.

Useful existing tools (suggest once if not installed): mattpocock/skills `prototype`, Anthropic `frontend-design`, Google Stitch for screen drafts; design resources in `references/design-resources.md` (check licences).

Versions

RepositoryScoreLicence
This page · Vamp-pire/hype-me95MIT
leanstacks/lstk100MIT
A7-studio/jobmigo100MIT
akillness/jeo-skills100MIT
runwield/runwield100MIT
yuan88964/RelayStudio100MIT
mathews-tom/armory100MIT
deaconb28-lang/prototype100MIT
drvoss/everything-copilot-cli100MIT

Database snapshot: 13.11.2026 · package version a119e42f479a (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 100experiment-coach
    Experiment Coach — validates ideas before launch and runs growth experiments after it, from inside the chat. Use when someone wants to test an idea, product, feature, offer, price, message or channel before building or launching it; get real market feedback; plan customer…
  • 100brainstorm-experiments
    Designs cheap experiments with a pre-set pass bar (at least X% of Y will do Z) and picks the method. Use when validating an idea before building, testing willingness to pay, or choosing between fake door, landing page and A/B test.
  • 100game-prototype
    Runs hypothesis-driven prototyping: one question per prototype, a falsifiable hypothesis, the cheapest fidelity that answers it (paper, simulation, agent-built throwaway), a time box, observed playtests and a kill-or-promote call. Use for "prototype", "proof of concept", "test…
  • 100prototype-scope-definer
    Define the minimum buildable artifact that tests the core design hypothesis. Prevent over-building before validation. Use this skill when: (1) defining what a prototype must include to test the core hypothesis, (2) making build-vs-fake decisions (what must be real, what can be…
  • 100prototype-anything
    Help a user explore, develop and articulate their own game or game-system idea over a multi-turn conversation, and write a faithful design document when asked. Use when the user brings an observation, concept or design notes they want to turn into a game. Not for implementing…
  • 82idea-to-mvp
    Turns an idea note into a one-page spec and a small working prototype, then tracks it as a project. Use when I say build this idea, make an MVP, prototype something, or use my creative block on an idea from Ideas/.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.