| name | deploy |
| description | Builds multi-language Astro site, verifies output for all language versions, generates deployment scripts and Nginx configuration with language-aware routing. Supports English, French, Kiswahili. Use after all pages are built, as the final step. |
Deploy
Acknowledgement: Shared by Peter Bamuhigire, techguypeter.com, +256 784 464178.
Use when
- The task matches this domain: Builds multi-language Astro site, verifies output for all language versions, generates deployment scripts and Nginx configuration with language-aware routing. Supports English, French, Kiswahili. Use after all pages are built, as the final step.
- The user needs an implementation-facing skill rather than a general discussion.
Do not use when
- The prerequisite upstream context is missing and the task is not yet execution-ready.
- Another narrower skill is the clear better fit for the exact subtask.
Required inputs
- Project context, current files, and any constraints that affect implementation.
- Upstream artifacts produced by earlier skills when this skill is part of a pipeline.
Workflow
- Read only the relevant project inputs and preserved guidance before acting.
- Choose the smallest set of references needed for the current job.
- Produce the implementation, configuration, or guidance this skill owns.
- Validate that the result stays compatible with the rest of the repository workflow.
Quality standards
- Outputs must be implementation-ready and internally consistent.
- Preserve existing behavior unless the task explicitly requires a change.
- Avoid host-specific path assumptions so the skill remains portable.
Anti-patterns
- Do not hardcode
.Codex/skills or another single install path.
- Do not skip validation against upstream or downstream dependencies.
- Do not generate generic output that ignores the actual project context.
Outputs
- Implementation guidance, configuration, generated artifacts, or concrete follow-on steps.
Canonical CI Pipeline
Every project launched through this skill must run the canonical CI pipeline
at templates/ci/website.yml. Install it with:
bash .Codex/skills/scripts/install-canonical-ci.sh <project-path>
The pipeline runs 15 ordered steps (install → lint → unit → build → e2e
smoke → metadata audit → perf gate → a11y gate → visual QA → security gate
→ drift check → design quality score → deploy → post-deploy smoke →
rollback-ready hook). Any failure blocks deploy.
Troubleshooting: references/ci-troubleshooting.md.
Launch Decision Record
Every launch records an explicit decision on:
- Observability — per
observability/SKILL.md, RUM, error tracking,
analytics, and alerts are live; the project runbook at
project-log/runbook.md names the on-call operator.
- Experimentation infrastructure — a decision under
project-log/decisions/ captures whether experimentation is enabled
for this client, which platform is provisioned, and the quarterly
cadence commitment. An answer of "no experiments yet" is acceptable
and must still be recorded; silence is drift.
- Design quality score — every primary template has a filed report
under
reports/design-quality/ and the aggregate is ≥ 8/10.
- Runbook review — the runbook has been reviewed in the last 90
days and lists the current on-call operator and escalation path.
A launch without these four recorded is not permitted to ship.
References
- Start with
references/legacy-guidance.md when you need the preserved detailed instructions from the previous skill version.
- Use
references/qa-matrix.md for the minimum launch QA standard across responsive behavior, content, links, performance, accessibility, SEO, analytics, forms, and browsers.
- Use
references/pre-launch-verification-checklist.md immediately before production release.
- Use
references/deployment-runbook.md for supported hosting patterns, env management, DNS/SSL, and promotion flow.
- Use
references/rollback-runbook.md to prepare and execute rollback safely.
- Use
references/observability-baseline.md for uptime, form delivery, analytics, and issue-detection expectations.
- Use
references/post-launch-review-checklist.md for launch-day, 7-day, and 30-day review windows.
- Use
references/launch-communication-template.md to standardize live-site notifications and internal launch comms.
- Use
references/performance-gate.md for the canonical performance enforcement layer and its thresholds.
- Use
references/ci-troubleshooting.md for the step-by-step triage of the 15-step canonical pipeline.
- Use
references/africa-calibration.md for the 3G network profile, weight budget, and data-cost rules the gates apply.
- Read only the specific files under
references/ that match the current task instead of loading the whole directory.
- Use
templates/playwright-starter/ when a project needs a minimal E2E baseline for smoke, navigation, and form coverage.
- Use
templates/ci/website.yml (via install-canonical-ci.sh) as the inherited CI pipeline; edit the canonical file, never the downstream copy.
Notes
- Treat this
SKILL.md as the portable execution layer for both Codex and Codex.
- Preserve existing project behavior unless the current task explicitly requires a change.