agentleFS
Sign inStart free
milad-ahmdmilad-ahmd/agent-skills/skills/startup-idea-validator/SKILL.md

startup-idea-validator

Skill5 starsChanged 15 days agoMITView on GitHub

Stress-tests a startup or product idea before you build it. It runs critique from a skeptical investor, a target customer, and a competitor, researches existing competitors, ranks the riskiest assumptions, writes Mom-Test interview questions, designs cheap validation experiments with success thresholds, and gives a Go, Pivot, or Kill verdict. Use when the user shares a startup, app, SaaS, or side-project idea and asks whether it's good, how to validate it, or what could kill it.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask the user for anything missing: who, what problem, why now, and why them.

    line 13
  2. Distinctive:

    No invented statistics, market sizes or competitor revenue.

    line 51

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Startup Idea Validator

Kill bad ideas cheaply, and make good ones sharper.

1. Restate the idea as a testable hypothesis

[Customer segment] has [problem] that happens [frequency], currently solved by [alternative], and will [pay / switch / use] for [solution] because [unique insight]. Ask the user for anything missing: who, what problem, why now, and why them.

2. The panel (3 short critiques, in character)

  • 🦈 Skeptical VC: market size logic, moat, distribution, "why won't Google or OpenAI do this?"
  • 🙋 Target customer: "Do I actually have this problem? What do I use today? Would I pay, and how much?"
  • ⚔️ Competitor CEO: "How would I crush or copy this in 3 months?"

3. Market reality check

  • Competitors and alternatives, including "do nothing", spreadsheets and manual workarounds. Use web search if it's available and cite your sources. Otherwise ask the user.
  • Where competitors are weak (look in their reviews, see competitor-spy).
  • Market size only from cited sources, or a clearly labelled bottom-up estimate (customers × price × frequency).

4. Riskiest assumptions (ranked)

AssumptionTypeRisk if wrongEvidence today
Freelancers lose >2h/week on invoicingDesirability🔴 fatalnone

Types: desirability (do people want it?), viability (will they pay, can it make money?), feasibility (can we build and distribute it?).

5. Cheapest experiments (1–2 weeks each)

For the top 3 assumptions, pick from: 10 customer interviews (Mom Test), a landing-page smoke test with a waitlist, a concierge or manual MVP, pre-sales or letters of intent, a fake-door feature, or community posts. Define the success threshold before running the experiment. For example: "≥ 8 of 10 interviewees describe the problem unprompted" or "≥ 5% of landing-page visitors join the waitlist".

6. Mom Test interview script

8–10 questions about their past behaviour, not about your idea ("Tell me about the last time you…", "What did you try?", "What did it cost you?"). Avoid "Would you use…?".

7. Verdict

🟢 Go / 🟡 Pivot / 🔴 Kill, with a confidence level, the 2–3 reasons behind it, what evidence would change the verdict, and a 2-week action plan.

Rules

  • Be candid, but always constructive. Attack the idea, never the founder.
  • No invented statistics, market sizes or competitor revenue. Cite them or label them as estimates.
  • This is not financial or legal advice.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free