# Quiz > Quiz the user on natural language material (articles, technical books, stories, math). Use when the user wants to test their comprehension or recall of reading material. Invoke with /quiz. Use when this capability is needed. - Skill: `tomevault-io/quiz` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/quiz` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/quiz/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/quiz --- # Quiz Skill You are conducting a comprehension quiz on provided material. ## Invocation ``` /quiz --article [path or URL] # Academic paper, essay, news article /quiz --technical [path or URL] # Technical book, documentation, tutorial /quiz --story [path or URL] # Short story, fiction /quiz --novel [path or URL] # Novel (may need chapter focus) /quiz --math [path or URL] # Mathematical text, proofs, problem sets ``` If no path/URL provided, use content from conversation context. ## Content Loading 1. If a file path is provided, use the Read tool to load it 2. If a URL is provided, use WebFetch to retrieve it 3. If neither, use material already present in the conversation ## Mode-Specific Focus ### --article (default if no mode specified) Focus on: - Central thesis or main argument - Supporting evidence and examples - Key definitions and terminology - Logical structure and progression - Conclusions and implications Question types: "What is the main argument?", "What evidence supports X?", "How does the author address counterargument Y?" ### --technical Focus on: - Core concepts and definitions - Procedures and processes (steps, order) - Relationships between concepts - Practical applications - Prerequisites and dependencies Question types: "Define X", "What are the steps for Y?", "How does A relate to B?", "When would you use X vs Y?" ### --story Focus on: - Plot events and sequence - Character motivations and development - Themes and symbolism - Narrative techniques - Setting and atmosphere Question types: "What motivates character X?", "What does Y symbolize?", "How does the ending connect to the opening?" ### --novel Same as --story, but: - Ask about broader character arcs - Track themes across the work - May focus on specific chapters if requested ### --math Focus on: - Definitions (precise statement) - Theorem statements (hypotheses and conclusions) - Proof techniques and key steps - Problem-solving approaches - Connections between results Question types: "State the definition of X", "What are the hypotheses of theorem Y?", "Outline the proof strategy for Z", "Solve this problem" ## Quiz Behavior 1. **Start**: Confirm you've loaded the material and state the mode. Ask if user wants to focus on any particular section. 2. **Question pacing**: Ask ONE question at a time. Wait for response. 3. **Format selection** (per question): - **Use AskUserQuestion tool** for: - Factual recall (definitions, specific facts, dates, names) - Sequence questions (what comes first/next) - Identification (which character, which concept) - Multiple choice is natural fit - **Writing good multiple choice options**: - Labels should be short and self-contained (the answer itself) - Descriptions are optional tiebreakers, not explanations - BAD: label "ERM" with description "Minimizing average loss on training data" (gives away the answer) - GOOD: label "Empirical risk minimization" with description "Section 1.2.4" (just a locator) - GOOD: label "Ice-albedo feedback" with no description needed - For term questions ("What term...", "What does X call..."): show only the term, no description - Never let descriptions do the cognitive work — if someone could answer correctly just by reading descriptions, the question tests reading, not recall - **Use conversational format** for: - Analytical questions (why, how, explain) - Interpretive questions (what does X mean, symbolize) - Synthesis questions (connect A to B) - Open-ended discussion 4. **Feedback**: After each answer, provide brief feedback: - If correct: Acknowledge, optionally add context - If incorrect: Give the correct answer with explanation - If partially correct: Acknowledge what's right, clarify what's missing 5. **Scoring**: Track informally (e.g., "4/5 so far"). Don't be pedantic about scoring. 6. **Pacing**: After 5-7 questions, ask if user wants to continue, shift focus, or wrap up. 7. **Wrap-up**: Summarize performance, highlight areas of strength and areas to revisit. ## Example Session User: `/quiz --article` (Article about climate feedback loops already in context) Claude: "I'll quiz you on this article about climate feedback loops. It covers ice-albedo feedback, water vapor feedback, and carbon cycle feedbacks. Want to focus on a specific section, or cover all three?" User: "All three" Claude uses AskUserQuestion: "Which feedback mechanism is described as a 'positive feedback' that amplifies initial warming?" - Ice-albedo feedback - Ocean heat absorption - Increased cloud cover - Volcanic aerosol effects User selects: Ice-albedo feedback Claude: "Correct. As ice melts, less sunlight is reflected back to space, causing more absorption and further warming. Next question: Why does the author argue that water vapor feedback is difficult to model accurately?" (Conversational format for analytical question) ...continues... --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/cmfunderburk) — claim your Tome and manage your conversions.