Web product spec
LaunchKit — build & launch a complete website with Claude Code. A bundle of skills (planning, design, UI/UX, frontend, SEO, accessibility, legal) that works worldwide with first-class Israel support (Hebrew/RTL, IS 5568, Israeli privacy law).
npx -y skills add JoshBuildApp/launchkit --skill web-product-specAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 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.
What its author says it does
Copied from the file, not written here
Use at the very start of a website project to decide WHAT to build before writing any code — target audience, the single main goal, must-have pages, content, and success metric. Triggers: "I want to build a site but not sure where to start", "help me scope my website", "what pages do I need", "plan my web app", defining requirements or an MVP for a site.
SKILL.md
2.9 KB, as published. Nobody here has run it
Web product spec
Pin down what you're building and for whom before touching design or code. Output a one-page spec.
Global default
Answer these, in order. Don't skip.
- Who is it for? One primary visitor. ("Padel coaches in Tel Aviv", not "everyone".)
- What's the one job? The single most important thing a visitor should do. One primary goal per site. Everything else is secondary.
- What does success look like? A number you can measure. (Signups/week, bookings, leads, sales.)
- Must-have pages. The smallest set that does the job. A landing page + 2–3 others usually beats 12 pages.
- Content inventory. What real text, images, and proof (logos, reviews) do you actually have? No lorem ipsum — fake content hides real problems.
- One primary call-to-action (CTA). The button you want clicked most. Repeat it; don't dilute it with five competing buttons.
- Then — and only then — pick the stack. Static site, no-code, or a framework? Match it to scope and who maintains it, not hype.
Output: the one-page spec
- Audience · Main goal · Success metric
- Page list (with the goal of each page)
- Primary CTA
- Content you have vs. content you still need
- Chosen stack + why
🇮🇱 Israel
- Decide language: Hebrew only, English only, or both? This drives RTL and design from day one (see
web-i18n-rtl). - If you'll sell or collect personal data, the site will need Israeli legal pages and likely an accessibility statement — flag it now so it's not a surprise at launch (see
web-legal,web-accessibility). - Decide payment reality early if selling: Israeli gateways (Cardcom, Tranzila, Grow/Meshulam) vs. global (Stripe). Affects checkout design.
🔌 Power-ups (installed skills)
Pull these in when they fit (skip any you don't have):
brainstorming— explore intent and requirements before locking scope.writing-plans— turn the spec into a step-by-step build plan.prompt-pro— if building through an AI tool (Lovable, Cursor, v0), craft the build prompt.- 🇮🇱
israeli-product-launch(launching to the Israeli market),shabbat-aware-scheduler(timeline around Shabbat/chagim).
Checklist
- One named primary audience
- One main goal + one measurable success metric
- Minimal page list, each page has a purpose
- One primary CTA chosen
- Real content inventoried (have vs. need)
- Stack chosen after scope, with a reason
- Language(s) decided (and RTL flagged if Hebrew)