Module: ObsceneGpt::Prompts

Defined in:

lib/obscene_gpt/prompts.rb

Constant Summary

SYSTEM_PROMPT =

<<~PROMPT.freeze
  You are a content moderation AI that analyzes texts for obscene, inappropriate, or NSFW content.

  Your task is to determine if the given text contains:
  - Explicit sexual content
  - Profanity or vulgar language
  - Hate speech or discriminatory language
  - Violent or graphic content
  - Other inappropriate material

  You will be given a JSON array of texts.
  You will need to analyze each text and determine if it contains any of the above content.
PROMPT

SIMPLE_SCHEMA =

{
  type: "object",
  properties: {
    obscene: {
      type: "boolean",
      description: "Whether the text contains obscene content",
    },
    confidence: {
      type: "number",
      minimum: 0,
      maximum: 1,
      description: "A confidence score between 0 and 1",
    },
  },
  required: %w[obscene confidence],
  additionalProperties: false,
}.freeze

FULL_SCHEMA =

{
  type: "object",
  properties: SIMPLE_SCHEMA[:properties].merge(
    reasoning: {
      type: "string",
      description: "A reasoning for the classification",
    },
    categories: {
      type: "array",
      items: { type: "string", enum: %w[sexual profanity hate violent other] },
      description: "A list of categories that the text belongs to",
    },
  ),
  required: %w[obscene confidence reasoning categories],
  additionalProperties: false,
}.freeze