Content moderation
Check usernames, reviews, comments and chat messages for offensive language. You get a safe/unsafe verdict, a risk score and, if you want them, the list of offending words — so you can reject, mask or send the message to a moderator.
What it detects
- Profanity and slurs in Arabic and English
- Obfuscated words (spacing, symbols, repeated letters)
- A 0–1 risk score you can tune
- Optional list of the bad words found
API methods
Sample response
{
"data": {
"isML": true,
"text": "This is just a normal text",
"totalBadWords": null,
"riskScore": 0,
"isSafe": true,
"status": "success",
"executionTime": 120
}
}