What it is
Jevfanity is a TypeScript API deployed on Cloudflare Workers that detects profanity and other moderation categories in text. It is for developers who need a hosted or self-hosted moderation endpoint.
How it uses Jev
Jev evaluates one Noul yes/no question per moderation category (profanity, slur, harassment, threat, sexual) on the input text. The returned probabilities are compared against a threshold and level policy to set flagged and flagged_categories.
Primitives:noul
Technique worth stealing
Ask one focused yes/no question per category instead of one broad question, then combine probabilities with a threshold.
Try it
curl -X POST https://api.jevfanity.app/v1/moderate -H "Content-Type: application/json" -d '{"text":"your message here","level":"medium"}'
Evidence
Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.
- Jev-centricyes0.89
- Shows a System One patternno0.27
- Handles uncertaintyno0.05
- Measuredno0.05
- Runnableyes0.82
- Worth recommendingno0.29
- Model replicano0.04
- Problem scopescore on a 0–2 scale1.97
- About Jevyes0.98
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 20 Sept 2026.