Testing and QA

Testing and QA

Current generated test-file count: 2884 across 28 workspaces.

Workspace	Runner	Test files
@claw/shared-auth	vitest	1
@claw/shared-constants	vitest	3
@claw/shared-entitlements	vitest	8
@claw/shared-rabbitmq	vitest	0
@claw/shared-types	vitest	3
@claw/shared-utilities	vitest	37
claw-agent-service	vitest	13
claw-audit-service	vitest	21
claw-auth-service	vitest	96
claw-chat-service	vitest	151
claw-client-logs-service	vitest	6
claw-connector-service	vitest	25
claw-file-generation-service	vitest	7
claw-file-service	vitest	20
claw-frontend	vitest	474
claw-health-service	vitest	4
claw-image-service	vitest	12
claw-llamacpp-service	vitest	17
claw-memory-service	vitest	14
claw-ollama-service	vitest	18
claw-payment-service	vitest	110
claw-research-service	vitest	31
claw-routing-service	vitest	99
claw-server-logs-service	vitest	7
claw-workspace-service	vitest	96

Test architecture

• Vitest is the workspace unit/integration runner.
• Playwright covers browser/E2E flows.
• Coverage is risk-based, with high global floors and stricter branch expectations for critical pure logic.
• Persistence, logs, RBAC, plan tiers, regressions, security and real browser behavior are part of completion evidence.
• Flaky tests, fixtures, contract tests, RabbitMQ, database, accessibility, visual and security testing each have dedicated standards.

Canonical sources: testing/README.md · testing/testing-strategy.md · testing/coverage-policy.md · rules/49-qa-team-discipline-and-test-evidence.md.