startup.ai platform help — an AI startup builder / no-code app builder that turns a plain-language idea into a working app and deploys it live: an AI Builder that generates a first app version, a Live Preview, an Idea Evaluation score (Market / Problem / Revenue), a browsable Idea Browser of AI-evaluated ideas, and one-click Publish to a public URL with a custom domain. Priced as a free prototype tier and a credit-metered Pro plan that unlocks public deploy and private ideas; UI-only, no public API. Use when building an app or MVP from an idea on startup.ai, reading its Market/Problem/Revenue idea score, browsing the Idea Browser, deploying to a custom domain, understanding the free vs Pro credits, or asking whether it has an API. Note: an AI idea score and a shipped app are not validated demand. Do NOT use for the validate-before-building method or comparing idea-validation tools (use /sales-idea-validation), or a general smoke-test landing page (use /sales-funnel).
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Um comando direto ignora o prompt de revisão. Verifique a origem antes de executá-lo.
startup.ai platform help — an AI startup builder / no-code app builder that turns a plain-language idea into a working app and deploys it live: an AI Builder that generates a first app version, a Live Preview, an Idea Evaluation score (Market / Problem / Revenue), a browsable Idea Browser of AI-evaluated ideas, and one-click Publish to a public URL with a custom domain. Priced as a free prototype tier and a credit-metered Pro plan that unlocks public deploy and private ideas; UI-only, no public API. Use when building an app or MVP from an idea on startup.ai, reading its Market/Problem/Revenue idea score, browsing the Idea Browser, deploying to a custom domain, understanding the free vs Pro credits, or asking whether it has an API. Note: an AI idea score and a shipped app are not validated demand. Do NOT use for the validate-before-building method or comparing idea-validation tools (use /sales-idea-validation), or a general smoke-test landing page (use /sales-funnel).
argument-hint
[describe what you need help with in startup.ai]
license
MIT
version
1.0.0
tags
["sales","pre-launch","platform"]
startup.ai Platform Help
startup.ai (startup.ai) is an AI startup builder / no-code app builder — "turn your idea into
reality." You describe an idea in plain language and it builds a working first version of the app
(with a Live Preview that updates as you edit), scores the concept in an Idea Evaluation (an
overall score plus Market / Problem / Revenue sub-scores), lets you browse a directory of
AI-evaluated ideas (the Idea Browser), and one-click Publishes the app to a public URL with a
custom domain (Vercel-hosted, HTTPS). It is the rare tool in this space that spans evaluate +
discover + actually build & deploy in one product — but a generated score and a shipped app are not
validated demand, and it has no documented public API (a UI tool).
⚠️ Name disambiguation. This is startup.ai (the AI builder). Several similarly named tools
exist — stratup.ai is a separate AI idea generator, startupadeasai.com ("Startup Ideas AI")
is another. If the user means an idea-generator/discovery tool, that's an /sales-idea-validation
question, not this skill.
Step 1 — Gather context
If references/learnings.md exists, read it first for accumulated platform knowledge.
Ask only what you can't infer:
What do you want from startup.ai?
A) Build an app / MVP from an idea (AI Builder, Live Preview, editing)
B) Read the Idea Evaluation score (overall + Market / Problem / Revenue) and what to do with it
C) Browse the Idea Browser to find an idea worth building
D) Publish / deploy to a public URL or custom domain (and what the free vs Pro plan gates)
E) Get more useful output, or ask whether it has an API to automate/export
Do you already have a clear idea description? Build and score quality track the detail you give it.
Are you trying to decide whether to build, or to build? If it's the go/no-go, that's a
/sales-idea-validation question — route in Step 2.
Skip-ahead: if the user's prompt already contains enough context, go straight to Step 3.
Step 2 — Route or answer directly
If the user's question is about…
Route to
The validate-before-building method, or comparing idea-validation/builder tools across the market
/sales-idea-validation {question}
Building the actual smoke-test / landing page the demand test runs on (as a general strategy)
/sales-funnel {question}
Growing a pre-launch waitlist / capturing real demand once you test
/sales-audience-growth {question}
Content / marketing strategy for the launched app
/sales-content {question}
A different similarly named tool (stratup.ai, Startup Ideas AI, an idea generator)
/sales-idea-validation {question}
When routing, give the exact command: "This is a {domain} question — run: /sales-idea-validation {original question}"
Otherwise, answer startup.ai-specific questions using Step 3.
Step 3 — startup.ai platform reference
Read references/platform-guide.md for the full reference — the module lineup (AI Builder, Live
Preview, Idea Evaluation, Idea Browser, Publish) and what each produces, how the Idea Evaluation score
is structured, the free-vs-Pro plan split and best-effort credit economics, the
custom-domain/Vercel deploy path, how to frame the idea input for better build + score output, and the
no-public-API automation reality.
Answer using only the relevant section — don't dump the full reference.
Step 4 — Actionable guidance
Read the Idea Evaluation as a directional AI opinion, not demand. The overall score and its
Market / Problem / Revenue sub-scores are an LLM's read of the concept — a high score is not
proof anyone will pay. Keep the structured parts (which sub-score is weak, the market framing) to
sharpen the pitch, and take the go/no-go from a real behavior test — route it to
/sales-idea-validation (smoke test / pre-sale).
Name the build-first trap explicitly. startup.ai makes building + deploying so frictionless that
the founder can skip demand testing entirely — the exact anti-pattern. Tell the user a shipped app
is not a validated app (building is cheap now; demand evidence is the scarce thing). The upside:
the app it deploys can double as the smoke-test page — publish it, drive a little targeted
traffic, and read the conversion; that is the proof.
Frame a rich idea input — build and score quality track input depth. A one-line prompt yields a
generic app and a generic score; feed it a specific customer, the exact problem, and how it's
different, then re-run and compare. Treat the generated app as an editable first draft.
Model pricing as free-vs-Pro with a monthly credit budget, and flag it as best-effort. Whenever
pricing or "is it worth it" comes up, name the split: a free tier (prototype/build in the editor)
vs a Pro monthly plan (best-effort ~a set number of credits/mo) that unlocks public deploy
and private ideas. Credits meter generations/builds, so iterating on the same app draws them down —
tell the user to confirm current tiers, credit counts, and what each build costs at startup.ai.
There's no public API — don't plan an integration around it. If asked to automate or export,
say plainly there's no documented public API, no webhooks, no Zapier/Make, no MCP; it's a UI tool.
The deployed app has its own public URL, but the builder itself can't be scripted — copy outputs
manually, or use a tool built for programmatic generation, and automate real signals (deployed-page
analytics), not generated assets.
Disambiguate the name before answering. startup.ai (this builder) is easily confused with
stratup.ai and "Startup Ideas AI" (idea generators) — if the user means a discovery/generator tool,
route to /sales-idea-validation.
If you discover a gotcha or tip not in references/learnings.md, append it there with today's date.
Gotchas
Best-effort from research (2026-07) — the module lineup, credits, and pricing move; verify at startup.ai.
An Idea Evaluation score is not validated demand. Market / Problem / Revenue sub-scores are an AI
opinion; pair them with a real demand test (/sales-idea-validation).
A shipped app is not a validated app. startup.ai's whole pitch — build + deploy in minutes — makes
it easy to skip validation. Use the deployed page as the smoke test; measure the committed action.
Public deploy and private ideas are Pro-gated. The free tier builds a prototype; publishing to a
public URL / custom domain and keeping ideas private need the paid Pro plan (best-effort — confirm).
Builds/generations burn credits. Pro includes a monthly credit budget; iterating on the same app
consumes it, so re-runs aren't free — budget accordingly.
Name collisions. startup.ai (this builder) ≠ stratup.ai (idea generator) ≠ startupadeasai.com
("Startup Ideas AI"). Route generator/discovery questions to /sales-idea-validation.
No documented public API. It's a web tool; don't plan integrations around the builder.
Related skills
/sales-idea-validation — The tool-agnostic validate-before-building method + the full validator/builder landscape (use this to actually decide build-or-not)
/sales-funnel — Build the smoke-test / landing page the demand test runs on (startup.ai's deployed app can serve as one)
/sales-frederick — Frederick AI — a sibling AI co-founder whose agents also build & ship apps and landing pages (credit-metered)
/sales-sololaunch — SoloLaunch — a sibling AI startup builder that generates a name/logo/brand/timeline asset kit (no app build)
/sales-ideabrowser — IdeaBrowser — the paid idea-discovery catalog (startup.ai's Idea Browser is a lighter, bundled version)
/sales-audience-growth — Grow a pre-launch waitlist and capture real demand
/sales-do — Not sure which skill to use? The router matches any sales objective to the right skill. Install: npx skills add sales-skills/sales --skill sales-do -a claude-code
Examples
Example 1: "startup.ai scored my idea 7.8/10 — should I build it?"
User says: "startup.ai gave my shift-scheduling idea an overall 7.8 with Market 8, Problem 7, Revenue 6. Good to build?"
Skill does: Reframes the score as a directional AI opinion, not demand — keeps the useful signal (Revenue is the
weakest sub-score, so willingness-to-pay is the open question) and prescribes a real test: build the app,
publish it as a smoke-test page, drive a little targeted traffic to a pre-set conversion threshold, then
attempt a pre-sale. Routes the go/no-go to /sales-idea-validation.
Result: The founder gets a decision from stranger behavior, using the score only to sharpen where to probe.
Example 2: Build and deploy an MVP, then know it's worth it
User says: "I want to build my MVP on startup.ai and ship it to a custom domain — is that validation?"
Skill does: Explains the AI Builder + Live Preview + one-click Publish flow and that public deploy /
custom domain is Pro-gated, then separates shipped from validated — startup.ai will happily deploy an
app nobody wants. Recommends using the deployed page as the smoke test (measure a committed action),
and routes the demand decision to /sales-idea-validation and the traffic capture to /sales-audience-growth.
Result: The user ships fast but gates the "is it worth it" verdict on real conversion, not the deploy.
Example 3: Can I automate startup.ai via API? (developer/automation)
User says: "I want to generate and score 20 ideas from a script — what's startup.ai's API?"
Skill does: States plainly there's no documented public API, no webhooks, no Zapier/Make, no MCP —
it's a UI tool, so there's no supported way to batch builds or pull scores. Suggests copying outputs
manually or calling an LLM API directly for bulk scoring, and notes the thing worth automating is a real
demand signal (deployed-page analytics), not generated apps — pointing to /sales-idea-validation.
Result: The user avoids building on a non-existent API and picks a workable path.
Troubleshooting
The generated app or score feels generic
Symptom: The built app is barebones or the Idea Evaluation reads like it could fit any product.
Cause: The idea input was thin — startup.ai's build and scoring need detail to beat a generic model.
Solution: Re-run with a rich description — specific customer, the exact problem, how the solution
differs, current stage — and compare variants. Treat the app as an editable first draft, not a finished MVP.
Can't publish / no custom domain / hit a limit
Symptom: The Publish/deploy option is locked, the custom domain won't attach, or generations stopped.
Cause: Public deploy and private ideas are Pro-gated, and builds/generations consume a monthly
credit budget — the free tier builds a prototype but reserves publishing and private ideas for Pro.
Solution: If you'll actively build and ship, weigh the monthly Pro fee and its credit budget against
how much you'll iterate; confirm the current tiers, credit counts, and per-build cost at startup.ai.
"Is there an API?" / wrong startup.ai
Symptom: Looking for API docs to automate builds/scores, or confused between startup.ai and stratup.ai.
Cause: startup.ai has no documented public API (it's a UI tool); stratup.ai and "Startup Ideas AI"
are different products — idea generators, not this builder.
Solution: Don't build an integration around the builder — copy outputs manually or call an LLM API
directly. If the task is generating/discovering an idea, that's /sales-idea-validation. Automate real
demand signals (deployed-page analytics) instead.
