plugins/mvp-launch/skills/mvp-launch/SKILL.md
Web app MVP launch checklist knowledge. Auto-triggered when conversation involves "launch readiness", "MVP checklist", "production ready", "go live", or pre-launch verification. Provides the 10-point criteria for what counts as "done" in each area. Based on "Realistic MVP Launch Checklist (from building 30+ apps)".
npx skillsauth add ooiyeefei/ccc mvp-launchInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
A battle-tested 10-point checklist for web app MVP launches. Based on building 30+ apps.
Philosophy: "Don't overbuild. Just make it stable, usable, and something people can trust."
What counts as DONE:
Required Webhook Events:
customer.subscription.trial_will_endcustomer.subscription.createdcustomer.subscription.updatedcustomer.subscription.deletedinvoice.payment_succeededinvoice.payment_failedinvoice.createdcheckout.session.completedVerification: See references/stripe-testing.md
What counts as DONE:
Key insight: Browser resize mode lies. Real device testing is mandatory.
Bonus (not required):
What counts as DONE:
Anti-pattern: Don't require 10 fields before users see value.
What counts as DONE:
Good defaults:
What counts as DONE:
Nice to have (don't overbuild):
What counts as DONE:
Goal: Catch bugs before users notice them.
What counts as DONE:
Examples: Typeform, in-app widget, email link, GitHub Issues integration.
What counts as DONE:
If using external auth (Clerk, Auth0): Most of this is handled for you.
What counts as DONE:
.vercel.app, .replit.app)Why it matters: Trust. Users won't pay for myapp.replit.app.
What counts as DONE:
Good choices: Supabase, Neon, PlanetScale, Convex, AWS RDS.
Anti-pattern: Replit DB, SQLite in production, no backups.
| Score | Meaning | |-------|---------| | 10/10 | Fully implemented, production-ready | | 7-9/10 | Mostly complete, minor gaps | | 4-6/10 | Partial implementation, needs work | | 1-3/10 | Minimal/broken implementation | | 0/10 | Not implemented |
Overall (sum of 10 areas):
references/search-patterns.md - Grep patterns for discoveryreferences/stripe-testing.md - Stripe CLI verification commandsdevelopment
This skill should be used when the user asks to "share this HTML", "publish HTML", "get a link for this file", "share this report", "make this shareable", "upload this HTML", or wants to publish any HTML artifact for others to view. ALSO use it for collaborative review on an HTML doc/spec/report — triggers include "get feedback on this", "let reviewers comment", "collect feedback", "share for review", "let people annotate this", "synthesize the feedback", "converge the feedback", "what did reviewers say", "incorporate the comments", or "improve this from the feedback". Wraps Surge.sh for zero-cost hosting with guided privacy options, plus an embedded annotation + AI converge workflow.
testing
Decide whether your agent actually needs persistent memory, feedback loops, or closed-loop learning, then design the smallest thing that pays for itself. Use when the user says "add memory", "give my agent context management", "make my agent learn", "self-improving / closed-loop", "Reflexion / mem0 / Letta / MemGPT", "AriGraph", "agent memory architecture", "long-term memory for chatbot", "why does my agent keep forgetting / making the same mistake", "fine-tune from agent traces", or asks for a memory schema / experience store / reward model. Filters ruthlessly — most teams want a state cache, not memory + learning. Default position is scratchpad-only with a stateless agent shipped first.
tools
Prescriptive Q&A workflow for designing agentic pipelines, multi-model councils, sub-agent hierarchies, and tool-loop hardening for any domain. Use when the user asks to "design an agent", "design a multi-agent system", "should I use a council/debate", "build a [domain] review agent" (HAZOP, finance, tutorial, marketing, compliance, accounting), "real agency vs workflow", "how to add sub-agents", "AI for [domain] review", or names patterns like "orchestrator-worker", "evaluator-optimizer", "Magentic", "ReAct", "plan-and-execute", "handoffs". Walks the user through 12 stages one question at a time and emits a buildable design doc with citations. Do NOT use for general coding questions, single-shot prompt tuning, or bare "use Claude to do X" requests with no agency requirement.
tools
End-to-end User Acceptance Testing for web applications. Analyzes branch changes and specs to generate exhaustive test cases, sets up the local environment, executes tests via Playwright browser automation, and produces a pass/fail results report with screenshots and fix documentation. Use when the user says "run UAT", "test this feature", "UAT testing", "acceptance test", "test my branch", "generate test cases", or wants to verify a feature branch against its spec before merge.