OpenAIModerationMiddlewareOptions

Whether to check user input messages.

Whether to check model output messages.

Whether to check tool result messages.

How to handle violations.

• "error": Throw an error when content is flagged
• "end": End the agent execution and return a violation message
• "replace": Replace the flagged content with a violation message

OpenAI model to use for moderation. Can be either a model name or a BaseChatModel instance.

Moderation model to use.

Custom template for violation messages. Available placeholders: {categories}, {category_scores}, {original_content}