Quality Assurance & Testing Strategy

To maintain the high reliability of the VENI-AI platform, we employ a multi-layered testing strategy. Every satellite must adhere to these standards to ensure seamless integration within the Shell.

---

🧪 The Testing Tiers

1. Unit Tests

• Scope: Individual functions, services, and utilities.
• Backend: Vitest + Ignis mock utilities.
• Frontend: Vitest + React Testing Library.
• Target: 80%+ coverage for business logic.

2. Integration Tests

• Scope: Controller-to-Service and Service-to-Database flows.
• Backend: Supertest + PostgreSQL (Test Containers or local test DB).
• Frontend: Component integration testing with MSW (Mock Service Worker).

3. End-to-End (E2E) Tests

• Scope: Critical user journeys across Shell and Satellites.
• Tool: Playwright.
• Required Scenarios:
  • Login via Keycloak -> Redirect to Shell.
  • App switching via Launchpad.
  • File upload/download in Drive.
  • Leave request cycle in HRM.

---

🏗️ Technical Standards

Contract-First Development

• All inter-service communication (gRPC) must be tested against the Protobuf definitions.
• Changes to protos/ require a full regression suite run for affected satellites.

Dependency Injection (DI)

• Use the @inject decorator in Ignis v2 to decouple services.
• Always provide mock implementations for external dependencies (e.g., EmailService, S3Client) during unit tests.

---

🚀 Automated Validation (CI)

Our GitHub Actions pipeline runs the following checks on every pull request:

1. Linting: npm run lint (ESLint + Prettier).
2. Type Checking: tsc --noEmit.
3. Tests: npm test.
4. Build: npm run build (Ensures production bundle integrity).

Blocked Merges

Pull requests cannot be merged if any automated check fails or if coverage decreases significantly.