Admin Dashboard — Service Readiness
Status: populated Owner: Platform Engineering (Frontend) Last updated: 2026-04-18
1. Readiness Checklist
Code Quality
- TypeScript strict mode; zero
anyin production code - ESLint + Prettier passing, zero warnings
- No
console.login production code - All components < 300 lines; pages < 500 lines
Testing
- Unit + integration coverage ≥ 80%
- All E2E Playwright flows passing in CI
- Accessibility: zero axe-core violations (WCAG 2.1 AA)
Security
- Security headers configured (CSP, HSTS, X-Frame-Options)
- Cloudflare Access protecting admin subdomain
- No secrets in source code or Docker image
- SMPP passwords never logged or returned after creation
-
pnpm audit: no HIGH/CRITICAL vulnerabilities - CSRF protection on all state-changing forms
Operations
-
GET /api/healthreturns 200 - K8s liveness and readiness probes configured
- Sentry DSN active; sensitive data scrubbing verified
- Structured JSON logging; admin user ID in all log lines
- Prometheus metrics endpoint exposed
- Alertmanager rules configured
Performance
- LCP < 2.5s on internal network (Lighthouse)
- Dashboard initial SSR < 500ms p95
- Chart rendering < 200ms after data update
- No render-blocking resources
Documentation
- All 17 service docs populated
- Local dev setup verified by a second engineer
- Runbook updated: ops team knows how to access and use the dashboard
2. Service Level Objectives
| SLO | Target |
|---|---|
| Availability | 99.5% monthly (internal tool; lower than customer-facing) |
| Error rate | < 1% of requests result in 5xx |
| Dashboard poll latency | < 300ms p95 for /api/metrics BFF aggregation |
| TTFB (initial page) | < 1s p95 on internal network |