L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

workflow-launch-ready

workflow-launch-ready

Launch-preparation sweep for a new app or major release. Use when "prepare for launch", "launch week", "everything before going live", "is the app launch-ready?", or "pre-launch sweep". Working tree to a PR → workflow-release-prep.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: workflow-launch-ready description: > Launch-preparation sweep for a new app or major release. Use when "prepare for launch", "launch week", "everything before going live", "is the app launch-ready?", or "pre-launch sweep". Working tree to a PR → workflow-release-prep. license: MIT metadata: chain: "enhance-web-seo enhance-pwa audit-bundle-size audit-i18n workflow-quality-gate deploy-verify iterate-post-launch"

workflow-launch-ready — Full Launch Preparation

Degree of freedom: MIXED. Skip/condition judgment [HIGH freedom]; sweep order and quality-gate NO-GO block [LOW freedom — run exactly].

The pre-launch sequence. Run once, get the complete picture across SEO, PWA, performance, quality, and post-launch monitoring.

Local dirty-tree → merge-ready PR is workflow-release-prep, not this.

How to reason

  1. Scope — new app / major release, not a dirty-tree PR
  2. Sweep — which steps apply; which are honest skips
  3. Verdict — LAUNCH-READY / WITH CONDITIONS / NOT READY
  4. Handoff — day-1 loop scheduled; blockers named

Worked example

Scope: first public launch of a Next.js + Supabase app; single locale; PWA wanted. Sweep: SEO + PWA + bundle + quality-gate; skip i18n with a note. Verdict: NOT READY — quality-gate NO-GO on missing RLS on profiles. Handoff: fix RLS, re-run Gate 2, then deploy-verify; do not treat this as workflow-release-prep.

Self-critique before reporting

  • Not a PR-prep — local dirty tree was not routed here
  • Skips named — PWA/i18n skipped only when inapplicable, in the table
  • NO-GO blocks launch — quality-gate NO-GO is NOT READY, not "with conditions"
  • Right owner — merge-ready PR → workflow-release-prep; npm package → deploy-npm

Sweep sequence [LOW freedom — run exactly]

1. SEO          → enhance-web-seo    (meta, OG, JSON-LD, sitemap, canonicals)
2. PWA          → enhance-pwa        (manifest, service worker, offline, install)
3. BUNDLE       → audit-bundle-size  (first-load JS ≤ 200 KB goal)
4. i18n         → audit-i18n         (if multi-locale: natural copy, no hardcoded strings)
5. QUALITY GATE → workflow-quality-gate (red-team + security + perf + unit tests)
6. DEPLOY       → deploy-verify      (post-deploy smoke after the release)
7. ITERATE      → iterate-post-launch (close loop on day-1 production signals)

Skip steps 2 (PWA) and 4 (i18n) if not applicable — note the skip in the checklist.


Step 1: SEO (read enhance-web-seo) [HIGH freedom]

Read the enhance-web-seo skill and follow it.

Every public page needs: <title>, meta description, OG tags, canonical URL, structured data (JSON-LD), and a sitemap entry.


Step 2: PWA (read enhance-pwa) [HIGH freedom]

Read the enhance-pwa skill and follow it.

Required: Web App Manifest + service worker with offline fallback, and no installability errors in Chrome DevTools → Application → Manifest (Lighthouse 12 removed its PWA score).


Step 3: Bundle size (read audit-bundle-size) [HIGH freedom]

Read the audit-bundle-size skill and follow it.

Target: first-load JS ≤ 200 KB gzip. All routes lazy-loaded.


Step 4: i18n (read audit-i18n — skip if single locale) [HIGH freedom]

Read the audit-i18n skill and follow it.

Priority: no hardcoded user-facing strings, natural copy in all supported locales, correct date/number/currency formatting.


Step 5: Quality gate (read workflow-quality-gate) [LOW freedom — hand off]

Read the workflow-quality-gate skill and follow it.

This step runs test-red-team, audit-security, audit-performance, and test-unit internally. A NO-GO verdict from this step blocks launch.


Step 6: Deploy & smoke (read deploy-verify) [LOW freedom — hand off]

Read the deploy-verify skill and follow it.

Run immediately after the production deploy. Confirm no new Sentry errors, schema is healthy, and critical flows work.


Step 7: Day-1 iteration loop (read iterate-post-launch) [LOW freedom — hand off]

Read the iterate-post-launch skill and follow it.

Schedule this 24–48 hours after launch. Pull first real-user signals and prioritize the first fix pass.


Launch checklist output [LOW freedom — run exactly]

## Launch Readiness — [App] — [Date]

### Verdict: LAUNCH-READY / LAUNCH WITH CONDITIONS / NOT READY

| Step | Status | Open items |
|------|--------|------------|
| SEO | ✅ / ⚠️ / ❌ | [list] |
| PWA | ✅ / ⚠️ / ❌ / skipped | [list] |
| Bundle size | ✅ / ⚠️ / ❌ | [X KB] |
| i18n | ✅ / ⚠️ / skipped | [list] |
| Quality gate | GO / NO-GO / CONDITIONAL | [see gate report] |
| Deploy smoke | ✅ / ❌ | [notes] |
| Day-1 plan | scheduled / not yet | [date] |

### Blockers (must fix before launch)
[list]

### Post-launch backlog (fix within 48 h)
[list]
Skills enregistrées