Moderation Guard

Text moderation demo: one TypeSafe systemOne call screens 7 Noul hazards + 1 severity Score in parallel. Verdict computed in code via policy thresholds.

Run

bun install
cp .env.example .env.local   # add TYPESAFE_API_KEY (optional — mock mode without it)
bun dev

Open http://localhost:3000.

API

POST /api/moderate { "text": "...", "policy": "strict" | "permissive" }

• Max 1000 chars, 10 req/min per IP (in-memory; use Redis for multi-instance prod).
• Without TYPESAFE_API_KEY returns keyword-heuristic mock with mock: true.
• With key: live Jev (jev-latest) via @typesafe-ai/sdk, server-side only.

Deploy (Vercel)

vercel --prod
# set TYPESAFE_API_KEY in Vercel env vars

Tune

Edit lib/moderation.ts: HAZARDS, POLICIES, route(). Thresholds should be fit on labeled traffic, not taken as universal.