# Generating Practice Questions

Generate educational practice questions from lecture notes to test student understanding.

- Repository: https://github.com/datawhalechina/agent-skills-with-anthropic (1565 stars)
- Category: Testing
- License: see repository
- Last updated: 2026-08-27
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add datawhalechina/agent-skills-with-anthropic --skill generating-practice-questions -g -a claude-code -y`
Codex: `npx skills add datawhalechina/agent-skills-with-anthropic --skill generating-practice-questions -g -a codex -y`
Cursor: `npx skills add datawhalechina/agent-skills-with-anthropic --skill generating-practice-questions -g -a cursor -y`
Gemini CLI: `npx skills add datawhalechina/agent-skills-with-anthropic --skill generating-practice-questions -g -a gemini-cli -y`

## Description

Generate educational practice questions from lecture notes to test student understanding. Use when users request practice questions, exam preparation materials, study guides, or assessment items based on lecture content.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Practice Question Generator

Generate comprehensive practice questions from lecture notes to test student understanding of learning objectives and key concepts.

## Input 

**Supported formats**: LaTeX (.tex), PDF, Markdown (.md), plain text (.txt)

- **PDF**: Use `pdfplumber` for text extraction

- **LaTeX**: Read as text, strip preamble (everything before `\begin{document}`), preserve math environments (`$...$`, `\[...\]`, `\begin{equation}`, etc.)

- **Markdown/Text**

**Content to extract**:

1. **Learning objectives** - Usually at beginning: "After this lecture, you should be able to..." or may be in section: "Learning Outcomes","Objectives", "Goals". If absent, derive from main topics.
2. **Main topics** - Section headings, bold terms, definitions, algorithms.
3. **Examples** - Use for realistic scenarios in questions.

## Question Structure

Generate questions in this order:

1. **True/False** (one per learning objective, or 3-5 if no objectives) 
2. **Explanatory Questions** (3-5 covering main topics)
3. **Coding Question**  (1 algorithm implementation or concept simulation)
4. **Use Case** (1 realistic application)

For each question type, follow guidelines below, and never include answer key.

## Question Guidelines

### Type 1: True/False

Test factual understanding and common misconceptions.

**Coverage**:

- One per learning objective, or 3-5 covering main topics if no objectives

**Difficulty progression**:

- Start with 1-2 simple definitional questions
- Include 2-3 reasoning-based questions requiring concept application or testing relationships between concepts

**Quality criteria**:

- Unambiguous with one correct interpretation
- Clear language without complex nested clauses
- Answer directly found in lecture notes
- Wrong answer reveals common misconception

**Examples**:

The rest: https://github.com/datawhalechina/agent-skills-with-anthropic/blob/HEAD/6.Creating Custom Skills（自定义skills）/generating-practice-questions/SKILL.md