Profanity Filter Tester
Paste any message and see how a production AI moderation engine scores it, live, across profanity, hate, harassment, spam, PII, and more. Try leetspeak or unicode tricks and watch them get caught.
Test a message
Paste your own text, press test, and watch the standard-mode categories react instantly.
What this tool tests
- 13 moderation categories: profanity, hate and discrimination, harassment, threats, sexual content, self-harm, violence, dangerous content, PII, spam, and more, each with its own score and threshold.
- Evasion resistance: text is normalized before scoring, so leetspeak ("fvck"), unicode lookalikes, and zero-width character tricks do not slip through.
- Real production engine: this is the same ensemble that powers the Paxmod API in production, not a simplified demo.
Frequently Asked Questions
Is this profanity filter tester free?
Yes, free and no signup. It runs your message through the same production engine that powers the Paxmod API.
Can it catch disguised profanity?
Yes. The engine normalizes leetspeak, unicode lookalikes, and zero-width characters before scoring. Paste a disguised message above and see for yourself.
How do I add this filter to my own app or game?
The same engine is one REST call away. See the API docs or the AI profanity filter service page. Usage-based pricing from $0.25 per 1,000 messages.
Want this filter in your app?
One POST call adds real-time moderation to your chat, game, or community. Free to start, no credit card.