​
Event Overview

The media.moderated event is triggered when the moderation process completes for a media item. This includes both AI-based and manual moderation decisions.

​
Payload Structure

{
  "event": "media.moderated",
  "timestamp": "2024-01-15T14:35:00Z",
  "data": {
    "media_id": "123e4567-e89b-12d3-a456-426614174000",
    "app_media_id": "user-upload-123",
    "site_id": "98765432-dcba-4321-fedc-210987654321",
    "status": "rejected",
    "previous_status": "pending",
    "moderation_type": "ai",
    "moderated_at": "2024-01-15T14:35:00Z",
    "moderation_details": {
      "rule_violations": [
        {
          "rule_id": "550e8400-e29b-41d4-a716-446655440000",
          "rule_name": "Inappropriate Content",
          "confidence": 0.95,
          "details": "Adult content detected"
        }
      ],
      "ai_analysis": {
        "adult": 0.95,
        "violence": 0.10,
        "hate": 0.05
      }
    },
    "metadata": {
      "user_id": "abc-987",
      "gender": "male"
    }
  }
}

​
Payload Fields

​
event
string
required
The event type. Always media.moderated for this event.
​
timestamp
string
required
ISO 8601 timestamp of when the event occurred.
​
data.media_id
string
required
Unique identifier for the media item in Pixel Patrol.
​
data.app_media_id
string
Your application’s identifier for this media item.
​
data.site_id
string
required
The site ID associated with this media.
​
data.status
string
required
The new moderation status: approved, rejected, or pending_review.
​
data.previous_status
string
required
The status before moderation.
​
data.moderation_type
string
required
How the content was moderated: ai, manual, or rule.
​
data.moderated_at
string
required
Timestamp when moderation completed.
​
data.moderation_details
object
Detailed information about the moderation decision.
​
data.moderation_details.rule_violations
array
List of rules that were violated (if any).
​
data.moderation_details.ai_analysis
object
AI confidence scores for different content categories.
​
data.metadata
object
Custom metadata provided during submission.

​
Status Values

  • approved: Content passed moderation and is safe to display
  • rejected: Content violated one or more rules and should not be displayed
  • pending_review: Content requires manual review by a moderator

​
Example Implementation

app.post('/webhook', async (req, res) => {
  const signature = req.headers['x-pixelpatrol-signature'];
  
  // Verify signature
  if (!verifyWebhookSignature(req.body, signature)) {
    return res.status(401).send('Invalid signature');
  }
  
  const { event, data } = req.body;
  
  if (event === 'media.moderated') {
    switch (data.status) {
      case 'approved':
        // Make content visible
        await db.media.update({
          where: { id: data.app_media_id },
          data: { 
            status: 'published',
            visible: true
          }
        });
        break;
        
      case 'rejected':
        // Handle rejected content
        await db.media.update({
          where: { id: data.app_media_id },
          data: { 
            status: 'rejected',
            visible: false,
            rejection_reason: data.moderation_details.rule_violations[0]?.details
          }
        });
        
        // Notify user
        await notifyUser(data.metadata.user_id, 'Content rejected');
        break;
        
      case 'pending_review':
        // Flag for manual review
        await db.media.update({
          where: { id: data.app_media_id },
          data: { 
            status: 'under_review',
            visible: false
          }
        });
        break;
    }
  }
  
  res.status(200).send('OK');
});

​
Use Cases

  • Content Publishing: Automatically publish approved content
  • User Notifications: Inform users about moderation decisions
  • Content Removal: Remove or hide rejected content
  • Manual Review Queue: Route edge cases to human moderators
  • Analytics: Track approval/rejection rates and reasons