question-set

Write discussion or text-dependent questions at a target rigor (Socratic, close reading). Do NOT use for assessment items (use assessment-item).

SKILL.md

GitHub

question-set

Creates 3-5 discussion or text-dependent questions scaffolded from lower to higher cognitive rigor (Bloom/DOK). Includes question type labels and suggested follow-ups.

Input

{
  "objective": "Students will analyze the author's use of figurative language.",
  "grade": "8",
  "subject": "ELA",
  "target_rigor": "DOK 3",
  "question_count": 4,
  "text_title": "The Road Not Taken"
}

Output

{
  "tool": "question-set",
  "questions": [
    {"level": "DOK 1", "question": "What two paths does the speaker describe?", "type": "literal"},
    {"level": "DOK 2", "question": "How does the speaker's description of each path differ?", "type": "inferential"},
    {"level": "DOK 3", "question": "Why might the speaker say the choice 'made all the difference' — is this sincere or ironic? Use evidence.", "type": "analytical"},
    {"level": "DOK 3", "question": "How does the poem's figurative language shape the reader's understanding of decision-making?", "type": "evaluative"}
  ],
  "human_review_required": true
}

Do NOT use this atom for

  • Assessment items with scoring (use assessment-item)
  • Reading comprehension quizzes (use assessment-designer)
  • Questions without a text or topic anchor

Pipeline note

Follows references/method.md at the Generation step (question design). Output conforms to references/metadata-schema.md. human_review_required: true — question quality and rigor alignment must be teacher-verified.