| name | answer-key |
| description | Generate an answer key with scoring notes for a set of assessment items. Use this atom when assessment-designer needs answer keys after items are generated. Do NOT use for generating the items themselves — that is atom-assessment-item. Do NOT use for grading student responses. |
answer-key
Creates an answer key with correct answers and scoring notes for each assessment item. Used after atom-assessment-item generates the items. Includes point values and common error notes.
Input
{
"items": [
{"id": "q1", "stem": "What is the main idea of paragraph 2?", "type": "multiple_choice", "options": ["A", "B", "C", "D"]},
{"id": "q2", "stem": "Explain how the author uses imagery.", "type": "constructed_response"}
],
"standard_code": "ELA.5.R.2.1"
}
Output
{
"tool": "answer-key",
"answers": [
{"id": "q1", "correct": "B", "points": 1, "scoring_note": "Accept B only; A is a detail, not the main idea"},
{"id": "q2", "exemplar": "The author uses imagery such as... to create a sense of...", "points": 2, "scoring_note": "Full credit requires citing specific imagery AND connecting to effect"}
],
"human_review_required": true
}
Do NOT use this atom for
- Generating assessment items (use atom-assessment-item)
- Grading actual student responses
- Building rubrics (use atom-rubric-build)
Pipeline note
Follows references/method.md at the Generation step (answer key). Output conforms to protocol-layer/metadata-schema.md. human_review_required: true — answer keys must be verified by a content expert.
