Detalle del Skill
hinge-question-designer
Creates formative classroom diagnostic questions.
Revisar antes de usar
La revisión automática comprueba relevancia, no seguridad ni respaldo. Lee las instrucciones de la fuente antes de usar este Skill.
SKILL.md
Este extracto es una copia guardada durante la revisión. La fuente externa contiene la versión completa y actual.
---
# AGENT SKILLS STANDARD FIELDS (v2)
name: hinge-question-designer
description: "Design a diagnostic hinge question that reveals whether students understand enough to move on. Use when planning key checkpoints mid-lesson during explicit or direct instruction."
disable-model-invocation: false
user-invocable: true
effort: medium
# EXISTING FIELDS
skill_id: "questioning-discussion/hinge-question-designer"
skill_name: "Hinge Question Designer"
domain: "questioning-discussion"
version: "1.0"
evidence_strength: "strong"
evidence_sources:
- "Wiliam (2011) — Embedded Formative Assessment: hinge questions as real-time diagnostic tools"
- "Christodoulou (2017) — Making Good Progress? The future of Assessment for Learning"
- "Black & Wiliam (1998) — Assessment and classroom learning"
- "Sadler (1989) — Formative assessment and the design of instructional systems"
- "Haladyna et al. (2002) — A review of multiple-choice item-writing guidelines for classroom assessment"
input_schema:
required:
- field: "concept_being_taught"
type: "string"
description: "The specific concept or skill the hinge question should diagnose understanding of"
- field: "student_level"
type: "string"
description: "Age/year group"
- field: "lesson_context"
type: "string"
description: "What has been taught so far in this lesson — the hinge question checks whether students understood this before moving on"
optional:
- field: "known_misconceptions"
type: "array"
description: "Specific misconceptions the teacher wants the distractors to target"
- field: "student_profiles"
type: "array"
description: "From context engine: prior assessment data, common gaps"
- field: "response_method"
type: "string"
description: "How students will respond — e.g. mini-whiteboards, fingers (A/B/C/D), clickers, hands up"
- field: "time_constraint"
type: "string"
description: "Maximum time for students to answer — hinge questions should be answerable in under 2 minutes"
output_schema:
type: "object"
fields:
- field: "hinge_question"
type: "object"
description: "The question stem, answer options, and correct answer"
- field: "diagnostic_key"
type: "object"
description: "What each answer choice reveals about student understanding — the misconception each distractor targets"
- field: "decision_guide"
type: "string"
description: "What to do based on class response patterns — when to proceed, re-teach, or differentiate"
- field: "design_rationale"
type: "string"
description: "Why the question and distractors are designed this way"
chains_well_with:
- "checking-for-understanding-protocol-designer"
- "explicit-instruction-sequence-builder"
- "practice-problem-sequence-designer"
- "retrieval-practice-generator"
teacher_time: "3 minutes"
tags: ["hinge-question", "formative-assessment", "misconceptions", "diagnostic", "checking-understanding"]
---
# Hinge Question Designer
## What This Skill Does
Designs a single, carefully crafted multiple-choice hinge question — a diagnostic question asked at a critical point in a lesson (the "hinge") to determine whether students have understood the key concept well enough for the teacher to move on. Unlike standard multiple-choice questions, every wrong answer in a hinge question is a carefully designed distractor that targets a specific, known misconception — so the teacher can tell not just WHO doesn't understand, but WHAT they don't understand and WHY. The output includes the question, the diagnostic key (what each answer reveals), and a decision guide (what to do based on class response patterns). AI is specifically valuable here because designing effective hinge questions requires simultaneously: identifying the critical concept, predicting the most common misconceptions, crafting distractors that would attract students holding each misconception but NOT stLeer la fuente completa en GitHub (abre una página externa)