Skip to main content

Testing

Environments​

Offline (step-by-step + full session)​

AUTH_ADAPTER=mock
CATALOG_ADAPTER=yaml
LLM_ADAPTER=mock
QUERY_ADAPTER=mock
SESSION_STORE=memory

Live QueryCraft auth + data​

AUTH_ADAPTER=querycraft
QUERY_ADAPTER=querycraft
QUERYCRAFT_ENABLED=true
QUERYCRAFT_URL=https://querycraft.ncrts.pro
QUERYCRAFT_TIMEOUT_MS=15000

Live OpenAI​

LLM_ADAPTER=openai
OPENAI_ENABLED=true
OPENAI_API_KEY=sk-...
OPENAI_MODEL_CLASSIFY=gpt-4o-mini
OPENAI_MODEL_INVESTIGATE=gpt-4o
OPENAI_MODEL_VERIFY=gpt-4o-mini

Restart after .env changes (npm run dev rebuilds docs then starts the API).

Smoke checklist​

  1. GET /api/steps — all core steps ready
  2. GET /api/health — adapters match env
  3. Pipeline curls for auth → catalog → classify → query
  4. Sessions — POST /api/advisor/sessions
  5. Redoc — confirm request/response examples

Automated tests​

npm test
npm run typecheck

Includes unit tests for auth/nlp/catalog/classify and HTTP e2e for pipeline + advisor sessions (mock).