Web App Deployment & Launch Fix Skill

About this skill

What it does

• Debugs deployment errors by working directly from your build/deploy logs
• Analyzes build logs to identify the true root cause, not just the first symptom
• Reviews environment variable setup across Production, Preview, and Development environments
• Reviews hosting platform setup: build command, install command, output directory, runtime version
• Checks build command and output directory configuration against your framework's actual behavior
• Troubleshoots domain and DNS configuration issues, including propagation vs. misconfiguration
• Reviews CI/CD pipeline stages (install → build → test → deploy) for misconfigurations
• Runs production readiness checks across database connections, API routes, and serverless functions
• Creates a tailored launch checklist based on your specific stack and integrations
• Builds a rollback and safety plan before any production-facing fix goes live

What is included

• A structured deployment debugging workflow covering stack identification through root-cause analysis
• A professional deployment framework covering build behavior, environment variables, package manager consistency, runtime versions, and more
• A ready-to-use launch checklist covering build, environment, functionality, domain, and safety checks
• A full set of output templates: Deployment Debug Report, Build Error Analysis, Environment Variable Review, Hosting Platform Setup Checklist, Launch Readiness Report, CI/CD Fix Plan, DNS/Domain Troubleshooting Report, and Post-Launch Smoke Test
• Three complete worked examples: a Vercel Next.js module resolution error, a Netlify output directory mismatch, and a Supabase environment variable/API connection issue
• Built-in safety rules that prevent secret exposure and enforce safe, staged production fixes

How to use it

1. Tell the skill your hosting platform (Vercel, Netlify, Cloudflare Pages, Render, Railway, Supabase, or GitHub Actions) and your app's stack/framework.
2. Describe what's failing — a build error, a blank page, a 404, a broken environment variable, or a domain/DNS issue.
3. When asked, paste your exact build or deploy log, and list the names (not values) of any relevant environment variables.
4. Review the root-cause analysis, which will clearly separate confirmed facts from assumptions.
5. Follow the step-by-step fix plan, testing in a preview/staging deployment first when your platform supports it.
6. Use the tailored launch checklist to confirm the app is genuinely ready to go live.
7. Run the post-launch smoke test after deploying to production to confirm everything works end-to-end.

Examples

"My Next.js app is connected to Vercel and auto-deploys from GitHub. The last deploy shows 'Build Failed' with an error I don't fully understand. It was working fine two days ago and I haven't changed much — just added a Supabase client for a new feature. Can you help me figure out what's wrong and get this deployed?"

**Root Cause (Hypothesis — pending log confirmation):** Based on the timeline, the most likely cause is a missing or misconfigured Supabase environment variable in the Production environment on Vercel, since the change coincides with adding a new Supabase client. This is not yet confirmed — the exact build log is needed to verify.

**Diagnostic Steps:**
1. Paste the full, exact Vercel build log so the specific failing step can be identified.
2. Confirm the names of any new environment variables the Supabase client requires (e.g. `NEXT_PUBLIC_SUPABASE_URL`, `SUPABASE_SERVICE_ROLE_KEY`) — values are not needed, only names.
3. Confirm whether those variables have been added to Vercel's Production environment settings, not just stored locally in `.env.local`.

**Build/Configuration Checks:**
- Confirm the build command and output directory haven't changed.
- Confirm the Supabase package was added to `package.json` and the lockfile was updated and committed.
- Confirm no local-only environment file is required at build time.

**Environment Variable Review:**
| Variable | Required | Set in Production? |
|---|---|---|
| `NEXT_PUBLIC_SUPABASE_URL` | Yes | To confirm |
| `SUPABASE_SERVICE_ROLE_KEY` | Yes (server-only) | To confirm |

**Fix Plan:**
1. Add any missing environment variables to Vercel's Production environment settings.
2. Confirm the service role key is not prefixed with `NEXT_PUBLIC_` — it must remain server-only.
3. Trigger a new deployment and monitor the build log for the same error.
4. If the build succeeds, test the Supabase-dependent feature directly in the deployed environment.

**Safety Notes:** Never paste secret values into logs, chats, or commits. If a service role key may have been exposed at any point, rotate it in the Supabase dashboard.

**Final Launch Checklist:**
- [ ] Build passes on Vercel after fix
- [ ] All Supabase environment variables present in Production
- [ ] Service role key confirmed server-only
- [ ] Supabase-dependent feature manually tested post-deploy

**Post-Launch Smoke Test:**
- [ ] Homepage loads over HTTPS
- [ ] New Supabase-powered feature loads real data
- [ ] No console errors on affected pages
- [ ] Existing features unaffected by the change

Known limitations

- This skill cannot access your hosting dashboard, private repository, live build logs, or environment variable values directly — it can only work from what you paste or describe.
- It will not invent log output, error messages, or platform settings on your behalf, and it will ask you to provide exact details when needed.
- Diagnoses given without full log access are clearly labeled as hypotheses rather than confirmed conclusions.
- For highly platform-specific or account-specific issues (billing limits, regional outages, provider-side incidents), you may also need to check your hosting provider's status page or support channel directly.

FAQ

Rate this skill

Comments