What it is
A Cloudflare Worker that checks text and usernames for profanity, returning a boolean verdict. It is aimed at trust-and-safety use cases and can be called from other Workers.
How it uses Jev
Jev answers one or two typed Noul questions per request: literal profanity in text, and for usernames also phonetic or look-alike disguised profanity. The Worker takes the max probability across questions, compares it to a threshold in code, and returns is_profane plus raw probabilities.
Primitives:noul
Technique worth stealing
Ask narrow typed questions and turn calibrated probabilities into a boolean verdict in code.
Try it
npm install; npx wrangler dev; open the printed local URL for Swagger UI.
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-centricyes0.96
- Shows a System One patternyes0.81
- Handles uncertaintyunclear0.55
- Measuredno0.06
- Runnableyes0.93
- Worth recommendingunclear0.59
- Model replicano0.03
- Problem scopescore on a 0–2 scale1.93
- About Jevyes0.99
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 21 Sept 2026.