agentsclimarketplace

Ship checklist

Skill pagoda111king/claude-code-pack/template/.claude/skills/ship-checklist

Claude Code config pack · 11 agents · 10 skills · 10 commands · 4 hooks · CLAUDE.md template · Solo Founder OS

Install
npx -y skills add pagoda111king/claude-code-pack --skill ship-checklist

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 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

Pre-launch checklist. Use before shipping a product or feature to external users. Solo Founder edition (not enterprise), trimmed to under 20 items to prevent real blockers that could sink you.

SKILL.md

5.0 KB, as published. Nobody here has run it

Pre-Launch Checklist · Solo Founder Edition

When to Use

  • Product MVP ready for first user trials
  • Before a major feature release
  • Anything that goes in front of customers

How to Use

Claude runs through each item and outputs ✅/⚠️/❌ + one line of evidence. Final verdict: GO / NO-GO.

Code & Security (7 items)

  1. ⬜ code-reviewer ran, no CRITICAL/HIGH unresolved or tracked
  2. ⬜ No hardcoded secrets: grepped for api_key|password|token|secret, no leaks
  3. ⬜ Dependency scan: npm audit / pip audit shows no high/critical vulnerabilities
  4. ⬜ .env not in git: .env in .gitignore, with .env.example template
  5. ⬜ Errors don't leak internals: production error messages contain no stack/SQL/path details
  6. ⬜ Production DB backup: at least one automated backup mechanism
  7. ⬜ HTTPS: all external interfaces enforce TLS

Functionality & UX (5 items)

  1. ⬜ Core happy path manually tested 3 times: once per different browser/device
  2. ⬜ 3 edge case tests: empty input / oversized input / disconnect-reconnect
  3. ⬜ Loading/Error states: every async operation has UI fallback, no frozen white screen
  4. ⬜ Mobile usable: if not a desktop tool, run through on iPhone
  5. ⬜ Destructive actions have confirmation: delete/charge/send must require second confirmation

Data & Compliance (3 items)

  1. ⬜ Privacy policy + Terms of Service: required if collecting user data (open-source templates OK)
  2. ⬜ ICP filing (if deployed in China + public-facing): unregistered sites get pulled
  3. ⬜ User data export/delete flow: meets basic data rights expectations

Payments (if charging, 4 items)

  1. ⬜ Payment full test: success / failure / cancel / duplicate submission each tested once
  2. ⬜ Refund/cancel path: user can self-serve or has a clear channel
  3. ⬜ Invoice/receipt: both business and individual options available
  4. ⬜ Webhook idempotency: Stripe/WeChat Pay callbacks won't double-credit

Operations & Marketing (4 items)

  1. ⬜ Error monitoring: at least Sentry or self-hosted logging + alerting
  2. ⬜ Uptime monitoring: UptimeRobot / BetterStack or similar
  3. ⬜ Customer support channel: at least one of WeChat blog / email / WeChat group, and email must be reachable
  4. ⬜ Launch materials: landing page + demo GIF/video + social feed copy + WeChat blog post (can summon @marketing)

Exemptions (explicitly skippable)

If the product is a pure internal tool / single-client custom build:

  • Can skip: 13, 14, 21 (case-by-case)
  • Cannot skip: 1-6, 8-12, 16-19 (if payments involved)

If it's a free open-source tool:

  • Can skip: 13, 14, 16-19
  • Cannot skip: 1-12, 20, 22, 23

Skipping requires a written reason, not default skip.

Output Format

# Pre-Launch Check: [Product/Feature Name]
Date: YYYY-MM-DD

## Item-by-Item Check

### Code & Security
1. code-reviewer ran —— ✅ Last run 2026-04-25, no CRITICAL, HIGH fixed
2. No hardcoded secrets —— ✅ Grepped, only .env.example has placeholders
3. Dependency scan —— ⚠️ 1 medium vulnerability (lodash), doesn't affect main flow, fix next release
4. ...

### Functionality & UX
8. Core happy path —— ❌ Only tested on Chrome, Safari not tested
...

## Status Summary

| Category | Total | ✅ | ⚠️ | ❌ | Skipped |
|---|---|---|---|---|---|
| Code & Security | 7 | 6 | 1 | 0 | 0 |
| ...

## Verdict

⚠️ **NO-GO · 2 issues to fix**:
1. #8 Safari not tested —— estimated 30 min
2. #20 Error monitoring not configured —— estimated 1 hour

Re-run this checklist after fixes. ETA: tonight.

GO Criteria

  • All CRITICAL items (1, 2, 4, 12, 16, 17) must be ✅
  • No more than 3 ❌ on other items, each with a clear "why it's OK to ship first" reason
  • Every ⚠️ has a ticket / next-release plan

Don'ts

  • Don't fake ✅ to hit a deadline (your code-reviewer sub-agent will catch it)
  • Don't replace items with "I'll fix it later" — either write a ticket or give an explicit skip reason
  • Don't let perfectionism block shipping — Solo Founders win through iteration, not first-version perfection

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.