# Co Learning Rubric Designer

> Creates analytic rubrics with performance level descriptors and criterion weighting. Use when DESIGNING evaluation criteria, creating scoring rubrics, defining performance levels, or weighting assessment criteria.

- Skill: `nahisaho/co-learning-rubric-designer` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add nahisaho/co-learning-rubric-designer`
- Raw SKILL.md: https://api.skillmd.com/api/skills/nahisaho/co-learning-rubric-designer/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: nahisaho (https://skillmd.com/u/nahisaho)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/nahisaho/co-learning-rubric-designer

---


# Rubric Designer

Creates analytic rubrics with performance levels, descriptors, and weighting.

## Use This Skill When

- Creating analytic rubrics for assignments or assessments.
- Defining performance level descriptors (Exemplary → Beginning).
- Weighting criteria to reflect objective priorities.
- Ensuring assessment criteria match stated learning objectives.

## Workflow

1. Receive objectives and activities from prior phases.
2. Define criteria derived from learning objectives.
3. Create 3–5 performance levels with distinct descriptors.
4. Assign criterion weights reflecting objective priorities.
5. Save rubric to `results/rubric.md` using asset template.

## Deliverables

- `results/rubric.md`: analytic rubric with levels and weights.
- `report.md`: rubric design rationale and alignment notes.

## Quality Gates

- [ ] Performance levels have distinct, non-overlapping descriptors.
- [ ] Criteria are observable and directly assessable.
- [ ] Weighting reflects learning objective priorities.
- [ ] Rubric criteria trace back to stated objectives.

If any gate fails: revise descriptors, adjust weights, and re-validate alignment.

## Gotchas

- 「優秀」レベルと「良好」レベルの記述子が曖昧に重なりやすい。具体的な行動指標で差別化すること
- 基準の重み付けは目標の優先度と一致させること。全基準を均等配分にしないこと
- 「発展途上」レベルを単に「良好の劣化版」として記述しないこと。学習段階に応じた独自の特徴を記述する

## Validation Loop

1. Draft rubric with criteria and performance levels
2. Check: distinct descriptors, observable criteria, weights aligned
3. If any check fails → revise descriptors or adjust weights
4. Rubric ready for feedback-analyzer input