You are an expert in natural language processing and explainable artificial intelligence.
Your task is to analyze textual examples and discover interpretable, human-understandable
features that can differentiate between {n_classes} text categories:
{class_list}

You must reason purely from the textual content itself, without assuming any predefined
feature types, and describe the distinguishing properties as if explaining them to a
human researcher.

I am providing you with example texts. Each text belongs to one of the {n_classes} categories
listed above, but you are NOT told which text belongs to which category.

Do not use the category names or paraphrases of them as feature names or
values. Describe the property itself instead, in general terms rather than
names or phrases taken from the examples.

Your task is to:
  1. Examine all texts carefully and propose key textual features that appear to
     systematically vary across subsets of the texts and together distinguish all
     {n_classes} categories in a classification model.
  2. Focus on characteristics meaningful to a human reader: differences in intent,
     narrative structure, linguistic choices, communicative function, level of abstraction,
     specificity, urgency, or discourse patterns.
  3. Each feature must capture a different property of the text. Do not propose two
     features that would almost always take related values.
  4. Some categories will resemble each other much more closely than others.
     Make sure the feature set also separates the closest pairs, not only the
     obvious groupings.
  5. Express each feature as a short snake_case attribute name.
  6. For each feature, provide a brief explanation of why it could be discriminative
     across the {n_classes} hidden groups.
  7. For each feature, provide 3-6 possible values: short snake_case strings
     without filler words, mutually exclusive, together covering the realistic
     range. If a feature cannot apply to some of the categories, include an
     explicit value for that case (for example not_applicable) in the list.
     These exact values will later be used to label new texts, so choose
     values you can apply consistently.
  8. Provide at least {min_features} distinct features.
  9. Respond only with valid JSON. Do not wrap it in markdown code fences and do not include any explanation outside the JSON.

Output format:
{{
  "proposed_features": [
    {{
      "feature": "feature_name_1",
      "description": "why it separates the {n_classes} groups",
      "possible_values": ["value1", "value2", "value3"]
    }}
  ]
}}
