Patient Portal Service — Service Readiness

Status: populated Owner: TBD Last updated: 2026-04-18 Companion: Service Template · docs/standards/DEFINITION_OF_DONE.md

1. Readiness Gate Checklist

All items must be green before the service is promoted to production.

Architecture & Design

• SERVICE_OVERVIEW.md reviewed and approved by platform architect
• DOMAIN_MODEL.md reviewed; aggregates and invariants confirmed
• API_CONTRACTS.md finalized; OpenAPI spec generated from controllers
• SECURITY_MODEL.md reviewed by security team
• FHIR resource surface reviewed by interop team

Implementation

• All 17 service docs populated (no stubs remaining)
• NestJS service scaffolded per hexagonal architecture (domain/, application/, infrastructure/, presentation/)
• Drizzle ORM schema matches DATA_MODEL.md
• PostgreSQL migrations checked in under src/infrastructure/migrations/
• RLS policies applied to all PHI tables
• Outbox relay worker implemented and tested
• Inbox deduplication implemented (NATS consumer group + inbox table)
• Feature flag ehr.portal entitlement check wired on all endpoints

Testing

• Unit test coverage ≥ 85%
• Integration test coverage ≥ 80%
• tenant-isolation.spec.ts green
• outbox.spec.ts green
• inbox.spec.ts green
• Pact consumer contracts for web + mobile registered and verified
• E2E patient login + MFA flow passing
• E2E lab results view with release policy passing
• E2E proxy delegation flow passing

Security

• SMART on FHIR scope enforcement verified for all 14 scopes
• MFA mandatory enforcement confirmed in Keycloak patient realm
• JWT ACR check on sensitive actions (export, account delete) verified
• Proxy scope escalation prevention tested
• PHI not appearing in application logs (log redaction verified)
• PHI not included in AI prompts (prompt builder test green)
• Release policy enforcement: unreleased results excluded in tests
• Security review sign-off from security team

Observability

• OpenTelemetry traces emitting to Tempo
• Prometheus metrics scraping active
• Portal Overview dashboard deployed in Grafana
• All P1 alerts configured in Alertmanager
• Runbooks linked from alert definitions

Operations

• GET /health liveness and readiness endpoints respond
• Kubernetes deployment manifests reviewed
• HPA configured (CPU + RPM triggers)
• Pod disruption budget configured (min 1 available)
• Rollout strategy validated (zero-downtime rolling)
• Environment variables documented in .env.example

Compliance

• GDPR export capability verified (POST /v1/portal/export tested)
• Access log completeness verified for all audit event types
• Data residency: PostgreSQL deployed in AFG region
• MoPH data governance sign-off obtained

---

2. Go/No-Go Criteria

Category	Must-pass	Should-pass
Testing	tenant-isolation, outbox, inbox, SMART scope tests	All E2E flows
Security	RBAC matrix, MFA enforcement, PHI log redaction	Security pen test
Observability	Alerts configured, traces active	Dashboard published
Compliance	GDPR export working, access log complete	MoPH sign-off