Zero to Launch · Aug 5 – Oct 14, 2026
28 skills,
free to take.
Every mechanical piece of the 6-session program, packaged as skill files. Plain markdown with frontmatter — run them in Claude Code, the Codex extension, or anything else that reads a skills directory. No account, signup or expiry.
Install
The archive contains a .claude/skills/ directory and nothing else, so it drops into an existing project without touching anything already there.
curl -fsSLO https://www.shipai.club/skills/zero-to-launch.zip
unzip -o zero-to-launch.zip -d .
rm zero-to-launch.zipPrefer one session's worth? Every session below has its own bundle. Each skill also links to its raw SKILL.md.
By session
/ask-santosSKILL.mdFind the right skill — Zero to Launch or Day One — for what you're actually trying to do. Use when someone describes a go-to-market problem in their own words, asks "where do I start", "what should I run", or doesn't know which skill applies.
/gtm-mapSKILL.mdMap a product onto a funnel you can actually measure and call the B2C/B2B fork. Use when the go-to-market plan feels shapeless.
/funnel-auditSKILL.mdFind which funnel stage loses the most people in absolute terms. Use when traffic exists but conversion doesn't.
/channel-planSKILL.mdDesign one acquisition channel with a test, a budget and a go/no-go number. Use when choosing where to spend the first effort.
/paid-testSKILL.mdDecide whether paid is worth trying yet and size the smallest honest test. Use before opening an ad account.
/community-planSKILL.mdDesign community as an instrumented funnel stage. Use when "start a Discord" is the plan and nobody has said what it's for.
/analytics-wireSKILL.mdInstrument a funnel end to end and verify every event really fires. Use when "we didn't really track that" is the honest answer.
/bi-agentSKILL.mdBuild an agent that reads your analytics on a schedule and reports what moved in plain English. Use once events are verified.
/icp-builderSKILL.mdDefine a B2B ICP tight enough to write one email to, and build the named account list. Use before any outbound.
/outbound-sequenceSKILL.mdWrite cold outbound email that gets replies, plus the follow-up cadence and the graceful no. Use once the account list exists.
/funnel-metricsSKILL.mdChoose the one metric per funnel stage that will actually get measured. Use after the funnel exists, before wiring anything.
/bi-agentSKILL.mdBuild an agent that reads your analytics on a schedule and reports what moved in plain English. Use once events are verified.
/site-structureSKILL.mdPlan a marketing site: sections, order, and the job each one does. Use before writing any page copy.
/landing-copySKILL.mdWrite marketing site copy from the positioning brief and apply it to the actual page. Use once the section order exists.
/site-scaffoldSKILL.mdScaffold or extend a Next.js marketing site — routes, metadata, sections. Use when the copy needs somewhere to live.
/og-imageSKILL.mdGenerate Open Graph images so shared links preview properly. Use once routes exist, before anyone shares one.
/perf-passSKILL.mdMeasure Core Web Vitals and fix only what moves a number. Use when the site is deployed and feels slow.
/content-mapSKILL.mdWork backwards from what the ICP searches to the first twenty pages worth writing. Use before drafting any article.
/programmatic-pagesSKILL.mdDesign a templated page set that is genuinely useful rather than a doorway. Use when one page shape times real data covers many queries.
/article-draftSKILL.mdDraft one article from the content map worth publishing under your own name. Use when the map names an unwritten page.
/social-auditSKILL.mdRead every profile a stranger finds when they look you up and find what contradicts your positioning. Use before a launch.
/social-profileSKILL.mdRewrite one social profile from the positioning brief — handle, bio, banner, pinned post, links. Use once per platform.
/social-automationSKILL.mdBuild a pipeline that drafts and queues social posts from the content map. Use when posting depends on remembering to.
/paid-basicsSKILL.mdPlan a performance-marketing push — structure, angles, targeting, daily loop — against a CAC ceiling. Use once paid has cleared the gate.
/paid-testSKILL.mdDecide whether paid is worth trying yet and size the smallest honest test. Use before opening an ad account.
/unit-economicsSKILL.mdModel CAC, LTV, churn and payback, and derive the most you can pay for a customer. Use before any paid spend.
/pricing-modelSKILL.mdPick or change a pricing model and check it against how customers actually get value. Use when price and value have drifted apart.
/launch-checklistSKILL.mdVerify everything that breaks a launch — site, metadata, analytics, links — by testing it. Use on launch day and hackathon weekend.
Questions people actually ask
Do I have to attend to use these?
No. They're published so anyone can run the process during the program or years from now.
Do they only work in Claude Code?
No. They're markdown files with frontmatter. Anything that reads a skills directory can run them.
Can I change them?
Yes, and you should. They're opinionated on purpose; edit them to fit the product and workflow.
What's the catch?
None. Ship AI is a free community meetup in Phoenix.
Ran one and it did something odd?
Say so in Discord. These files improve based on what happens when people actually run them.