From super-knowledge
Validates scoring changes against psychometric standards. Use when the user modifies scoring weights, formulas, thresholds, bands, normative references, or composite scoring. Triggers: scoring change, weight change, formula change, scoring review, validate scoring, threshold change, band change, composite score, normative reference.
How this skill is triggered — by the user, by Claude, or both
Slash command
/super-knowledge:scoring-reviewerThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Validates proposed scoring changes against psychometric standards. Catches changes that would compromise measurement validity, reliability, or fairness.
Validates proposed scoring changes against psychometric standards. Catches changes that would compromise measurement validity, reliability, or fairness.
Before reviewing, read these domain files for reference:
skills/psychometric-advisor/scoring.md — scoring methods and principlesskills/psychometric-advisor/reliability.md — reliability implications of scoring changesskills/psychometric-advisor/validity.md — validity implications, DIF considerationsThe user provides one of:
Apply each check. Mark PASS, FAIL, or N/A with specific reasoning.
Not all checks require the same level of human oversight:
docs/solutions/, surface it explicitly: "This conflicts with [prior solution]. Has the context changed?"## Scoring Review
### Change Summary
[1-2 sentence description of what changed]
### Checklist Results
| # | Check | Status | Notes |
|---|-------|--------|-------|
| 1 | Composite reliability | PASS/FAIL/N/A | [reasoning] |
| 2 | Weight justification | PASS/FAIL/N/A | [reasoning] |
| ... | ... | ... | ... |
### Critical Issues
[Issues that must be resolved before shipping]
### Warnings
[Issues that should be addressed but don't block shipping]
### Verdict
**APPROVE** / **APPROVE WITH CONDITIONS** / **REJECT**
[1-2 sentence summary with the key reason]
npx claudepluginhub aibilitycz/superpowered-toolkit --plugin super-knowledgeProvides patterns for weighted scoring systems, quality gates, rubrics, and threshold-based decisions to evaluate code reviews, proposals, or artifacts.
Designs structured scoring rubrics with explicit criteria, performance scales, and descriptors for consistent quality assessment and reduced subjective bias.
Audits a proposed assessment for construct validity, reliability, and alignment to learning objectives. Use when reviewing or quality-assuring assessments before deployment.