audit-questions

Audit and rewrite multiple-choice questions for knowledge mapping domains.

6|2|Updated Nov 13, 2025
One-click install
npx skills add https://github.com/ContextLab/mapper --skill audit-questions
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audit-questions
Source: https://github.com/ContextLab/mapper/tree/main/.claude/skills/audit-questions
Command: npx skills add https://github.com/ContextLab/mapper --skill audit-questions

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill systematically audits and improves the quality of multiple-choice questions used in knowledge mapping applications, ensuring they accurately test conceptual understanding rather than trivia or recognition.

Core Features & Use Cases

  • Quality Auditing: Detects and flags common question flaws like excessive length, recognition-based testing, poor distractors, guessable answers, keyword leakage, cultural bias, and definition-as-question formats.
  • Iterative Rewriting: Automatically rewrites flagged questions to meet strict quality and difficulty criteria.
  • Use Case: A curriculum designer can use this Skill to audit a set of 50 physics questions, identify those that are too easy or poorly phrased, and have them automatically rewritten to be more effective learning tools.

Quick Start

Use the audit-questions skill to audit questions for the biology domain.

Frequently Asked Questions about audit-questions

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit multiple-choice questions to fix weak distractors and keyword leakage?

Auditing multiple-choice questions involves detecting quality issues like weak distractors, guessability, and keyword leakage, then iteratively rewriting the flagged questions until they meet strict standards for testing conceptual understanding.

What is recognition testing in assessment design and how do I eliminate it?

Recognition testing occurs when questions assess trivia identification rather than conceptual understanding. You eliminate it by auditing questions for definition-based formats and excessive length, then rewriting them to evaluate deeper knowledge application.

How do I automatically rewrite poorly phrased questions for knowledge mapping domains?

To automatically rewrite poorly phrased questions for knowledge mapping, you run an iterative quality audit that accepts a domain name, flags structural flaws like cultural bias, and generates revised questions meeting strict difficulty criteria.

Can I use an automated question quality audit for a large set of physics or biology questions?

Yes, an automated question quality audit can process large sets of physics or biology questions by systematically identifying poorly phrased items and rewriting them into more effective learning tools for knowledge mapping.

What are the common flaws in multiple-choice question design that compromise knowledge mapping?

Common multiple-choice question flaws compromising knowledge mapping include excessive length, recognition testing, weak distractors, guessable answers, keyword leakage, cultural bias, and definition-as-question formats.