Protect your community without building a full moderation team. The
moderate_content and
pii_detect task types
review text, images, profiles, and files — flagging violations with confidence scores and explanations.
Detect hate speech, harassment, spam, misinformation, and policy violations in posts and comments.
Flag NSFW content, violence, fake documents, and copyright violations in uploads.
Review user profiles for fake identities, impersonation, and fraudulent listings.
Find and redact social security numbers, credit cards, passports, and personal data from documents.
Apply custom content policies with AI + human escalation for borderline cases.
Track violation trends, category breakdowns, and false-positive rates over time.
curl -X POST https://api.crowdsorcerer.dev/v1/tasks \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"type": "moderate_content",
"input": {
"content": "User-generated text or image URL here...",
"content_type": "text",
"policies": ["hate_speech", "harassment", "spam", "adult"],
"threshold": 0.7
}
}'
# Response output:
{
"verdict": "flag",
"confidence": 0.92,
"violations": [
{ "policy": "harassment", "confidence": 0.92, "excerpt": "..." }
],
"recommended_action": "remove"
}Set a confidence threshold. Items below it are automatically routed to our 18-type worker network for human review — with full audit trails, appeals workflow, and mediator escalation for disputes.
500 free credits on signup. No credit card required.