APIVerveScreen user content before it goes live.
Pass every submission through profanity, sentiment, and spam checks to decide — automatically — what publishes, what's held for review, and what's dropped.
- Profanity Filter1 cr
- Sentiment Analysis5 cr
- Spam Detector5 cr
User-generated content is a moderation firehose. Reviewing everything by hand doesn't scale; publishing everything is a brand risk. A lightweight gate handles the obvious cases so humans only see the gray area.
A publish / hold / reject decision for every submission, so moderators spend their time only on genuinely borderline content.
How the calls chain together
Every response below is a live sample from the actual API — nothing mocked.
Profanity FilterPOST /v1/profanityfilter1 crFilter profanity
Detect and mask profane language, and get a cleaned version of the text back.
Sample response
{
"isProfane": true,
"filteredText": "Today is so **** hot! Why the **** would anyone go outside?",
"mask": "*",
"trimmed": false,
"profaneWords": 2
}
Sentiment AnalysisPOST /v1/sentimentanalysis5 crRead the tone
Score sentiment to route hostile or highly negative content toward review.
Sample response
{
"comparative": 0.25,
"sentimentText": "positive",
"sentiment": 3,
"isPositive": true,
"isNegative": false,
"normalizedScore": 0.3
}
Spam DetectorPOST /v1/spamdetector5 crCatch spam
Flag likely spam, including disposable emails and blacklisted IPs in the submission.
Sample response
{
"likelySpam": true,
"isDisposableEmail": false,
"isIPBlacklisted": false,
"ipDetails": {
"country": "IN",
"region": "DL"
},
"parsed": true
}