Enterprise Content Moderation API
Enterprise-Grade Safety: Our API is designed for mission-critical applications with 99.9% uptime SLA, enterprise support, and real-time moderation at scale.
The Content Moderation API is a powerful, enterprise-grade solution designed to automatically detect and filter harmful content across text, images, videos, and audio with military-grade precision and sub-second response times.
Text Safety
Detect toxicity, hate speech, harassment, and spam across 100+ languages.
API Logic Flow & Decision Process
Intelligent Processing: Our API uses sophisticated decision trees and machine learning algorithms to determine the most appropriate moderation approach based on your parameters.
How the API Processes Your Request
Step 1: Parameter Validation
- Required:
content_type,moderate,api_key - Optional:
threshold(default: 0.7),custom_instruction - If any required parameter is missing →
400 Bad Request
Step 2: Content Type Processing
- text: Direct text analysis
- text_custom_instruction: Uses your custom moderation rules
Step 3: Moderation Categories Decision
Scenario A: "standard" only
Uses 16 predefined categories: Finance, Health, Religion & Belief, Legal, Toxic, Insult, Public Safety, War & Conflict, Illicit Drugs, Profanity, Politics, Derogatory, Violent, Sexual, Death Harm & Tragedy, Firearms & Weapons
Scenario B: "standard" + custom words
Uses all 16 standard categories PLUS your custom moderation topics with custom words you defined (max 50 total)
Step 4: AI Analysis Process
- Confidence Scoring: Determines confidence scores with respect to 16 standard moderation types
- Text Parsing: Parses text to match words and phrases against 16 standard types and any additional custom ones added by user
- Custom Instructions: Moderates text based on custom specific instructions which allows more granular control over moderation than by setting moderation types only. You can e.g. instruct our API to keep specific words even though they may belong to one of moderation types, or remove specific custom words even though they do not belong to one of moderation types
Step 5: Response Generation
- moderation: Confidence scores for standard moderation types (JSON)
- moderation_words: Specific phrases found matching your categories. Each matched phrase is represented in array as element [phrase, moderation type, start index of phrase in text, end index of phrase]
- moderation_custom_instruction: Custom analysis result (if applicable)
- status: 200 (success) or error code
Authentication
Secure Authentication: All API keys are encrypted at rest and in transit using AES-256 encryption with rotating keys.
API requests should be sent to the secure endpoint:
Base URL: https://www.contentmoderationapi.com/api/
A valid API key is required to access our enterprise API endpoints. API keys are available through our enterprise subscription plans. After obtaining a plan, log in to your dashboard to retrieve your unique API key.
Include your API key in all requests as a parameter:
api_key: your_api_key_here
Security Notice: Replace your_api_key_here with your actual API key. Keep your API key secure and never expose it in client-side code or public repositories.
Advanced Content Moderation
Submit a POST request with your content and specify which types of harmful content to detect. Our AI-powered engine analyzes content across multiple dimensions and returns detailed safety scores and classifications.
AI-Powered Detection: Our machine learning algorithms continuously improve detection accuracy, achieving 99.2% precision across all supported content types and languages.
Moderation Parameter Logic
Parameter Behavior: Understanding how the "moderate" parameter works is crucial for effective content moderation.
What happens with different "moderate" values?
Empty "moderate" Parameter
"moderate": ""
Result: API returns error
{
"error": "moderate parameter must be specified. Use 'standard' for default categories or provide comma-separated custom words (max 50)",
"status": 400
}
"standard" Only
"moderate": "standard"
Result: Uses 16 predefined categories
- Finance
- Health
- Religion & Belief
- Legal
- Toxic
- Insult
- Public Safety
- War & Conflict
- Illicit Drugs
- Profanity
- Politics
- Derogatory
- Violent
- Sexual
- Death, Harm & Tragedy
- Firearms & Weapons
"standard" + Custom Words
"moderate": "standard,betting,gambling,crypto"
Result: Uses all 16 standard categories PLUS your custom words
- All 16 standard categories
- + betting
- + gambling
- + crypto
Note: Maximum 50 total categories allowed
Complete API Parameters
| Parameter | Type | Required | Possible Values | Description |
|---|---|---|---|---|
| content | string | Required | Any text or URL | The text content to be analyzed for moderation. |
| content_type | string | Required | text, text_custom_instruction | Type of content to analyze. Supports text analysis and custom instruction moderation. |
| moderate | string | Required | "standard" or comma-separated custom words | Moderation categories. Use "standard" for 16 predefined categories, or specify custom words (max 50 total). |
| custom_instruction | string | Conditional | Any instruction text | Required when content_type is "text_custom_instruction". Your custom moderation instructions. |
| threshold | float | Optional | 0.0 to 1.0 | Confidence threshold for flagging content. Defaults to 0.7. |
| api_key | string | Required | Your API key | Your unique enterprise API key for authentication. |
Enterprise SDKs & Code Examples
Production-ready code examples for seamless integration into your content platforms. All examples include error handling, retry logic, and security best practices.
Comprehensive Enterprise Example
The following example demonstrates our API analyzing potentially harmful content across multiple dimensions:
Text Moderation Request
curl -X POST -H 'Content-Type: application/x-www-form-urlencoded' \
-d 'content=This is some user-generated content to analyze for safety&content_type=text&moderate=standard,betting,gambling&api_key=your_api_key' \
'https://www.contentmoderationapi.com/api/predict'
Python Integration
import http.client
import urllib.parse
import json
import logging
# Configure enterprise logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
def moderate_content(content, content_type, moderation_types, api_key, threshold=0.7):
"""
Enterprise-grade content moderation with error handling and logging
"""
try:
conn = http.client.HTTPSConnection("www.contentmoderationapi.com")
payload = urllib.parse.urlencode({
'content': content,
'content_type': content_type,
'moderate': ','.join(moderation_types),
'threshold': threshold,
'api_key': api_key
})
headers = {
'Content-Type': 'application/x-www-form-urlencoded',
'User-Agent': 'Enterprise-Content-Moderation/1.0'
}
conn.request("POST", "/api/predict", payload, headers)
response = conn.getresponse()
data = response.read().decode("utf-8")
if response.status == 200:
result = json.loads(data)
logger.info(f"Content moderated - Status: {result.get('status', 'Unknown')}")
return result
else:
logger.error(f"API request failed with status {response.status}")
return None
except Exception as e:
logger.error(f"Content moderation failed: {str(e)}")
return None
finally:
conn.close()
# Enterprise usage example
user_message = "Visit the best betting site for amazing odds and win big money now!"
moderation_types = ['standard', 'betting'] # Standard categories + betting
result = moderate_content(user_message, 'text', moderation_types, 'your_api_key')
if result:
if result.get('status') == 200:
print("✅ Content analyzed successfully")
print("Moderation Confidence Scores:", result.get('moderation', 'N/A'))
print("Specific phrases found matching your categories:", result.get('moderation_words', []))
print("Each matched phrase is represented in array as element [phrase, moderation type, start index of phrase in text, end index of phrase]")
else:
print("⚠️ Content analysis failed")
print("Error:", result.get('error', 'Unknown error'))
else:
print("❌ Failed to connect to API")
import urllib.parse
import json
import logging
# Configure enterprise logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
def moderate_content(content, content_type, moderation_types, api_key, threshold=0.7):
"""
Enterprise-grade content moderation with error handling and logging
"""
try:
conn = http.client.HTTPSConnection("www.contentmoderationapi.com")
payload = urllib.parse.urlencode({
'content': content,
'content_type': content_type,
'moderate': ','.join(moderation_types),
'threshold': threshold,
'api_key': api_key
})
headers = {
'Content-Type': 'application/x-www-form-urlencoded',
'User-Agent': 'Enterprise-Content-Moderation/1.0'
}
conn.request("POST", "/api/predict", payload, headers)
response = conn.getresponse()
data = response.read().decode("utf-8")
if response.status == 200:
result = json.loads(data)
logger.info(f"Content moderated - Status: {result.get('status', 'Unknown')}")
return result
else:
logger.error(f"API request failed with status {response.status}")
return None
except Exception as e:
logger.error(f"Content moderation failed: {str(e)}")
return None
finally:
conn.close()
# Enterprise usage example
user_message = "Visit the best betting site for amazing odds and win big money now!"
moderation_types = ['standard', 'betting'] # Standard categories + betting
result = moderate_content(user_message, 'text', moderation_types, 'your_api_key')
if result:
if result.get('status') == 200:
print("✅ Content analyzed successfully")
print("Moderation Confidence Scores:", result.get('moderation', 'N/A'))
print("Specific phrases found matching your categories:", result.get('moderation_words', []))
print("Each matched phrase is represented in array as element [phrase, moderation type, start index of phrase in text, end index of phrase]")
else:
print("⚠️ Content analysis failed")
print("Error:", result.get('error', 'Unknown error'))
else:
print("❌ Failed to connect to API")
Complete Response Examples
Successful Analysis Response
{
"status": 200,
"moderation": "[{\"category\":\"Finance\",\"confidence\":0.87},{\"category\":\"Toxic\",\"confidence\":0.23},{\"category\":\"Health\",\"confidence\":0.12},{\"category\":\"Religion & Belief\",\"confidence\":0.05},{\"category\":\"Legal\",\"confidence\":0.34},{\"category\":\"Insult\",\"confidence\":0.19},{\"category\":\"Public Safety\",\"confidence\":0.08},{\"category\":\"War & Conflict\",\"confidence\":0.15},{\"category\":\"Illicit Drugs\",\"confidence\":0.03},{\"category\":\"Profanity\",\"confidence\":0.41},{\"category\":\"Politics\",\"confidence\":0.28},{\"category\":\"Derogatory\",\"confidence\":0.16},{\"category\":\"Violent\",\"confidence\":0.09},{\"category\":\"Sexual\",\"confidence\":0.07},{\"category\":\"Death, Harm & Tragedy\",\"confidence\":0.11},{\"category\":\"Firearms & Weapons\",\"confidence\":0.04}]",
"moderation_words": [
["money", "Finance", 45, 50],
["betting", "betting", 12, 19],
["win big", "gambling", 67, 73]
],
"total_credits": 10000,
"remaining_credits": 9987,
"processing_time_ms": 234
}
Custom Instruction Response
When using content_type: "text_custom_instruction":
{
"status": 200,
"moderation": "[{\"category\":\"Finance\",\"confidence\":0.45},{\"category\":\"Toxic\",\"confidence\":0.12},{\"category\":\"Health\",\"confidence\":0.08},{\"category\":\"Religion & Belief\",\"confidence\":0.03},{\"category\":\"Legal\",\"confidence\":0.22},{\"category\":\"Insult\",\"confidence\":0.05},{\"category\":\"Public Safety\",\"confidence\":0.04},{\"category\":\"War & Conflict\",\"confidence\":0.07},{\"category\":\"Illicit Drugs\",\"confidence\":0.01},{\"category\":\"Profanity\",\"confidence\":0.09},{\"category\":\"Politics\",\"confidence\":0.15},{\"category\":\"Derogatory\",\"confidence\":0.06},{\"category\":\"Violent\",\"confidence\":0.02},{\"category\":\"Sexual\",\"confidence\":0.03},{\"category\":\"Death, Harm & Tragedy\",\"confidence\":0.05},{\"category\":\"Firearms & Weapons\",\"confidence\":0.01}]",
"moderation_words": [],
"moderation_custom_instruction": "This content contains promotional language that may encourage financial risk-taking behavior. Consider adding disclaimer or warning labels.",
"total_credits": 10000,
"remaining_credits": 9986
}
Error Response Examples
Missing Required Parameter:
{
"error": "moderate parameter must be specified. Use 'standard' for default categories or provide comma-separated custom words (max 50)",
"status": 400
}
Invalid Content Type:
{
"error": "content_type must be one of: text or text_custom_instruction",
"status": 400
}
Enterprise Error Handling
Robust Error Handling: Our enterprise API provides detailed error responses with actionable guidance and 24/7 support escalation for critical issues.
All API responses include a status field. Status code 200 indicates successful processing. Our enterprise SLA guarantees comprehensive error reporting and rapid resolution.
HTTP Status Codes
| Status Code | Description | Resolution |
|---|---|---|
| 200 | Success: Content successfully analyzed. | ✅ Process the moderation results as needed. |
| 400 | Bad Request: Invalid content or parameters. | 🔧 Verify content format and required parameters. |
| 401 | Unauthorized: Invalid or missing API key. | 🔑 Check API key validity and permissions. |
| 403 | Quota Exceeded: Monthly request limit reached. | 💳 Upgrade plan or purchase additional credits. |
| 410 | Content Load Error: URL could not be loaded or text too short. | 📝 Verify URL accessibility or provide longer text. |
| 411 | Access Denied: Content blocked by security measures. | 🔒 Content may be behind authentication or firewall. |
| 429 | Rate Limited: Too many requests per minute. | ⏱️ Implement request throttling or increase rate limits. |
| 500 | Server Error: Internal processing error. | 🆘 Contact enterprise support for immediate assistance. |
24/7 Enterprise Support
Dedicated support team available around the clock for enterprise customers. Average response time: < 15 minutes for critical issues.
Contact SupportStatus & Monitoring
Real-time API status dashboard with performance metrics, uptime monitoring, and proactive incident notifications.
View Status