- Check text content for potentially harmful material
- Identify specific categories of harmful content
- Get confidence scores for moderation decisions
Available Endpoints
Moderations Check
Check content for potentially harmful or inappropriate material.Request Format
Request includes the input text to be moderated.Response Format
The response includes:- Overall moderation decision (flagged/not flagged)
- Category-specific flags (e.g., violence, hate speech, sexual content)
- Confidence scores for each category
- Highlighted sections of concerning content
Use Cases
- Preventing harmful content in AI responses
- Filtering user inputs for inappropriate material
- Ensuring compliance with content policies
- Creating safe AI interactions for all users
- Protecting brand reputation