Arbiter — Case management

A template-based screening and assessment tool for case-management teams. Admins author reusable assessment templates (sections, questions, scoring bands, conditional visibility); case workers at agencies run screenings against their clients, track live scores, submit locked records, and review aggregated reports.

Stack

• API: NestJS + Drizzle ORM + postgres-js
• Client: React + Vite + TypeScript + Recharts
• DB: PostgreSQL 16 (docker-compose)

Local dev

docker compose up -d
cd api && npm install && npm run db:migrate && npm run seed && npm run start:dev
cd web && npm install && npm run dev

• API on :3001
• Web on :5173 (proxies /api to :3001)
• Postgres on :5434

E2E tests

cd web && npm run test:e2e

Playwright covers: masthead + Admin/Clinical toggle, template list + publish/fork/archive, template authoring (create, sections/questions/options/bands/conditionals, save, publish), screenings ledger + filters + CSV export, screening detail (all three question types, conditional visibility, live scoring, per-answer notes, can-submit gating, submit lifecycle + lock), and the aggregate reports dashboard. Tests require the API + seed data; the suite's globalSetup reseeds the database at the start of the run.

Structure

arbiter/
  api/                NestJS REST API
  web/                React client (Admin + Clinical modes)
  docker-compose.yml

Notable design decisions

• Screenings snapshot their template at creation time; in-flight casefiles are unaffected by later template edits, forks, or archives.
• Published templates cannot be edited — fork to create a new draft version instead.
• Scoring is computed on read while a screening is in flight; final score + band are persisted on submission.
• Conditional questions are hidden when their dependency isn't satisfied, and skipped questions are excluded from the scoring denominator rather than counted as zero.
• Question-type polymorphism: true_false, multiple_choice, and likert share the same ScreeningAnswer shape with typed fields.