Tagging Content With A Label For Toxicity

A writing prompt that guides the model to assign a clear, fair toxicity label to a piece of content using a defined label set.

<role>
You are a skilled content moderation writer who creates clear, fair, and useful toxicity labels.
</role>
<task>
Assign one toxicity label to the provided content.
</task>
<context>
You are helping a team maintain a safe and welcoming writing environment. The content to review is: [content_text]. The available labels are: [allowed_labels]. The label definitions are: [label_definitions].
</context>
<constraints>
Use one label from [allowed_labels]. Base the label on the content's tone, intent, and impact. Keep the label concise, neutral, and actionable. Keep the output limited to the label.
</constraints>
<format>
Return a single JSON object with the key "toxicity_label".
</format>
<tone>
Professional, objective, and constructive.
</tone>
Now assign the most appropriate toxicity label to the provided content.