Idea validator
Use when the user has a product/feature idea and asks "should I build this", "is there a market", "validate this idea" — or is about to spend days building something unvalidated.From its SKILL.md
npx -y skills add dersi3/idea-validator-skill --skill idea-validatorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 26 days oldThe repository was created 26 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.7 KB, 625 tokens by cl100k_base, as published. Nobody here has run it
Idea Validator
Overview
A 30-minute evidence protocol that ends in BUILD / RESHAPE / SKIP — before any code exists. The goal is disconfirmation: actively look for reasons the idea fails, because the builder already believes it works.
Step 1 — State the bet (2 min)
Force one sentence: "{who} will pay {price} for {thing} because {pain}." If "who" is "everyone" or "because" is vague, stop — reshape until the sentence is falsifiable.
Step 2 — Demand evidence (15 min, web searches)
Hunt in this order:
- Are people already paying? Search for direct competitors and their pricing. Competitors are good news (proven wallet) unless the space is saturated with free options.
- Is the pain expressed unprompted? Search Reddit/HN/forums for the problem phrased by sufferers ("how do I…", "is there a tool that…", complaint threads). Save 3-5 verbatim quotes with links — these become marketing copy later.
- Is anything free "good enough"? The real competitor is usually a spreadsheet, a free tool, or doing nothing. Name it explicitly and answer: why would they switch?
- Can buyers be reached? Name 2-3 concrete channels where "{who}" congregates and whether they permit/have precedent for product posts.
Step 3 — Score honestly (5 min)
| Question | 0 | 1 | 2 |
|---|---|---|---|
| Wallet proof | nobody pays for anything like this | adjacent things sell | direct competitors charge money |
| Unprompted pain | found nothing | vague grumbling | verbatim "I need this" quotes |
| Free-alternative gap | free covers it fully | free is annoying | free genuinely can't do it |
| Reachability | no idea where buyers are | 1 plausible channel | 2+ proven channels |
| Buildable small | months to v1 | weeks | days |
8-10 → BUILD (smallest sellable version, presell if possible). 5-7 → RESHAPE (usually: narrower "who" or different "thing", rerun). 0-4 → SKIP — write the one-line kill reason and keep it; killed ideas recur.
Step 4 — The output
One page: the bet sentence, the evidence with links (especially verbatim pain quotes), the score table, the verdict, and — for BUILD — the smallest version sellable within days plus its first distribution channel.
Common mistakes
- Searching only for confirmation ("great idea for X" articles) instead of the pain.
- Treating friends' "I'd totally use that" as evidence. Only strangers' wallets and unprompted complaints count.
- Scoring 7 and building the original idea anyway instead of reshaping. The score is the decision, or the protocol is theater.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.