Edit Step4_q_board Template

Step4_q_board v2

Viewer

Test Case:

Template

Preview

History

System prompt (optional) User prompt (template)

You are analyzing the Questions section from an NSPE Board of Ethical Review case.

**QUESTIONS SECTION TEXT:**
{{ questions_text }}

**EXTRACTED CASE ENTITIES:**
{{ entities_text }}

{{ provisions_text }}

**TASK:**
Extract ONLY the Board's explicit questions (the actual questions posed to the Board).
These are the questions the case was asked to answer.

For each question:
1. **Question Text**: The verbatim question text
2. **Mentioned Entities**: Which entities from the case are referenced? Use exact labels from the list above.
3. **Related Provisions**: Which code provisions (if any) are mentioned?
4. **Reasoning**: What is this question really asking?

**OUTPUT FORMAT (JSON):**
```json
[
  {
    "question_number": 1,
    "question_text": "Was it ethical for Engineer A to accept the contract?",
    "question_type": "board_explicit",
    "mentioned_entities": {
      "roles": ["Engineer A"]
    },
    "related_provisions": ["II.4.e"],
    "extraction_reasoning": "Asks whether accepting the contract violated ethical duties."
  }
]
```

Extract ALL questions the Board was asked. Use EXACT entity labels from the lists above.

Loading preview...

Version History

No version history yet. Save changes to create versions.

Saved

1144 chars

Reference

Template Variables

Raw Questions section text from the case document

All nine entity types formatted as label plus short definition

CODE PROVISIONS EXTRACTED block (up to 10 provisions, 100-char text cap); empty string when none

Output Schema

format json_array

item_fields

JSON Wrapper

Appended to the prompt by the extractor:

IMPORTANT: Wrap your response as a JSON object with exactly two keys:
{"new_step4_q_board_classes": [...], "step4_q_board_individuals": [...]}
If there are no individuals to report, use an empty array for "step4_q_board_individuals".