The Moderation API checks text for policy violations — a free tool to keep harmful content out of your application.

How to Use It

response = client.moderations.create(
    input="User's message here"
)

result = response.results[0]
if result.flagged:
    # Block this content
    categories = result.categories
    print(f"Flagged: {categories}")

What It Detects

• Hate speech, harassment
• Violence and graphic content
• Self-harm content
• Sexual content
• Illicit activity

Best Practice

Run moderation on all user inputs before sending to the main model. It's free and adds important safety protection.

Reference:

Moderation API documentation

https://en.wikipedia.org/wiki/Special:Search?search=Moderation%20API