All skills
dmccreary avatar

/chapter-content-generator

@3ccea10

Generates detailed chapter content for an intelligent textbook — text, diagrams, MicroSims, and exercises at the appropriate Bloom's level. Use when a chapter's index.md exists with title, summary, and concept list.

Use this Skill: https://skilld.dev/gh/dmccreary/claude-skills/chapter-content-generator

This session only. Nothing lands on disk.

referencesblooms-taxonomy.md

≈3.6k tokens on demand. Your agent reads this file only when SKILL.md points to it.

<!-- Canonical copy. Do not fork into other skills; reference by path: $BK_HOME/skills/chapter-content-generator/references/blooms-taxonomy.md `$BK_HOME` points at the ibook-skills checkout and is agent-neutral. Do NOT use a path under `~/.claude/skills/`, `~/.codex/skills/`, or `~/.gemini/antigravity/skills/` — each is specific to one agent framework, and these skills run on all of them. If `$BK_HOME` is unset, the file is at `chapter-content-generator/references/blooms-taxonomy.md` relative to the skills root that contains the calling skill. -->

Bloom's Taxonomy (2001 Revision) — Canonical Reference

Shared by chapter-content-generator (learning objectives), faq-generator (question distribution), and quiz-generator (question levels). Merged from the former bloom-taxonomy-verbs.md and blooms-taxonomy-guide.md.

Part 1: Action Verbs by Level

Remember (L1)

list, define, recall, identify, name, recognize, locate, describe

Understand (L2)

explain, summarize, interpret, classify, compare, contrast, exemplify, infer

Apply (L3)

use, execute, implement, solve, demonstrate, calculate, apply, practice

Analyze (L4)

differentiate, organize, attribute, compare, contrast, examine, deconstruct, distinguish

Evaluate (L5)

judge, critique, assess, justify, prioritize, recommend, validate, defend

Create (L6)

design, construct, develop, formulate, compose, produce, invent, generate

Part 2: Writing Questions at Each Level

The guidance below was written for FAQ generation; the question starters, characteristics, and level distinctions apply equally to quiz questions.

The Six Cognitive Levels

1. Remember (Knowledge)

Definition: Recall facts, terms, basic concepts, and answers.

Cognitive Actions:

  • Define, duplicate, list, memorize, recall, repeat, reproduce, state

Question Starters:

  • "What is...?"
  • "What does [term] mean?"
  • "What are the components of...?"
  • "Can you list...?"
  • "What are the main...?"

FAQ Examples:

✓ "What is a Learning Graph?" ✓ "What does MicroSim mean?" ✓ "What are the five levels of intelligent textbooks?" ✓ "What is the ISO 11179 standard?"

Answer Characteristics:

  • Factual, straightforward
  • Draws from glossary or definitions
  • 50-150 words
  • May reference source for more detail

Target Distribution in FAQ: 20%


2. Understand (Comprehension)

Definition: Explain ideas or concepts; demonstrate understanding of facts and ideas.

Cognitive Actions:

  • Classify, describe, discuss, explain, identify, locate, recognize, report, select, translate

Question Starters:

  • "How would you explain...?"
  • "Why is [concept] important?"
  • "What is the difference between...?"
  • "How does [concept] work?"
  • "Can you summarize...?"
  • "What is the main idea of...?"

FAQ Examples:

✓ "Why are learning graphs important for textbook design?" ✓ "What is the difference between a glossary and an index?" ✓ "How does scaffolding improve learning outcomes?" ✓ "What's the relationship between Bloom's Taxonomy and question difficulty?"

Answer Characteristics:

  • Explanatory, contextual
  • May compare/contrast concepts
  • Often includes "because" or "in order to"
  • 100-250 words
  • Examples help clarify

Target Distribution in FAQ: 30%


3. Apply (Application)

Definition: Use information in new situations; solve problems using required skills or knowledge.

Cognitive Actions:

  • Apply, build, choose, construct, develop, implement, interview, make use of, organize, solve, utilize

Question Starters:

  • "How would you use...?"
  • "What examples can you find...?"
  • "How would you solve...?"
  • "How would you show...?"
  • "What approach would you use to...?"
  • "How would you apply...?"

FAQ Examples:

✓ "How do I create a learning graph for my course?" ✓ "What's the best way to use MicroSims in my chapters?" ✓ "How can I apply Bloom's Taxonomy to my quiz questions?" ✓ "When should I use an infographic vs. a MicroSim?"

Answer Characteristics:

  • Procedural, instructional
  • Step-by-step guidance
  • Practical examples
  • "How to" focus
  • 150-300 words
  • May include code or concrete steps

Target Distribution in FAQ: 25%


4. Analyze (Analysis)

Definition: Draw connections among ideas; determine how parts relate to overall structure or purpose.

Cognitive Actions:

  • Analyze, break down, compare, contrast, diagram, differentiate, distinguish, examine, experiment, identify, investigate, relate

Question Starters:

  • "What is the relationship between...?"
  • "What are the parts or features of...?"
  • "How does [X] relate to [Y]?"
  • "What evidence can you find...?"
  • "What is the underlying cause of...?"
  • "What patterns can you identify...?"

FAQ Examples:

✓ "What's the relationship between concept dependencies and learning order?" ✓ "How do the five levels of textbook intelligence build on each other?" ✓ "What factors determine whether a concept needs a MicroSim?" ✓ "How does prerequisite structure affect chapter organization?"

Answer Characteristics:

  • Analytical, investigative
  • Breaks down relationships
  • Identifies patterns or causes
  • May use diagrams or charts
  • 200-300 words
  • Multiple perspectives considered

Target Distribution in FAQ: 15%


5. Evaluate (Evaluation)

Definition: Justify a decision or course of action; make judgments based on criteria and standards.

Cognitive Actions:

  • Appraise, argue, defend, judge, select, support, value, critique, evaluate, assess, recommend

Question Starters:

  • "Which approach is best for...?"
  • "What are the trade-offs of...?"
  • "How would you decide between...?"
  • "What criteria would you use to...?"
  • "How would you prioritize...?"
  • "What are the strengths and weaknesses of...?"

FAQ Examples:

✓ "Which approach is better for teaching complex concepts: text or MicroSim?" ✓ "What are the trade-offs between comprehensive glossaries and brief definitions?" ✓ "How do I decide which concepts need prerequisite validation?" ✓ "What's the best balance between content depth and cognitive load?"

Answer Characteristics:

  • Evaluative, comparative
  • Discusses pros/cons
  • Uses criteria for judgment
  • May present multiple valid options
  • 200-350 words
  • Nuanced, considers context

Target Distribution in FAQ: 7%


6. Create (Synthesis)

Definition: Produce new or original work; reorganize elements into new patterns or structures.

Cognitive Actions:

  • Assemble, construct, create, design, develop, formulate, generate, plan, produce, propose, invent

Question Starters:

  • "How would you design...?"
  • "What would happen if...?"
  • "Can you create a...?"
  • "How could you improve...?"
  • "What's the best way to combine...?"
  • "How would you adapt [X] for [Y]?"

FAQ Examples:

✓ "How would I design a textbook for a completely new subject area?" ✓ "What's the best way to combine learning graphs with adaptive learning?" ✓ "How could I create an innovative assessment beyond traditional quizzes?" ✓ "How would you design a MicroSim to teach a highly abstract concept?"

Answer Characteristics:

  • Creative, generative
  • Proposes novel solutions
  • Encourages experimentation
  • Open-ended guidance
  • 250-400 words
  • Multiple approaches suggested

Target Distribution in FAQ: 3%


Target Distribution by Category

Different FAQ categories should emphasize different Bloom's levels:

Getting Started (60% Remember, 40% Understand)

Focus on basic facts and understanding course structure.

Example Questions:

  • "What is this course about?" (Understand)
  • "What are the prerequisites?" (Remember)
  • "How is the textbook organized?" (Understand)

Core Concepts (20% Remember, 40% Understand, 30% Apply, 10% Analyze)

Balance definitions with understanding and application.

Example Questions:

  • "What is a learning graph?" (Remember)
  • "Why are concept dependencies important?" (Understand)
  • "How do I create a concept taxonomy?" (Apply)
  • "What's the relationship between scaffolding and prerequisites?" (Analyze)

Technical Details (30% Remember, 40% Understand, 20% Apply, 10% Analyze)

Emphasize terminology and technical understanding.

Example Questions:

  • "What does ISO 11179 mean?" (Remember)
  • "How does the glossary validator work?" (Understand)
  • "When should I use cross-references?" (Apply)
  • "How do metadata standards improve consistency?" (Analyze)

Common Challenges (10% Remember, 30% Understand, 40% Apply, 20% Analyze)

Focus on problem-solving and troubleshooting.

Example Questions:

  • "Why is my learning graph showing cycles?" (Understand)
  • "How do I fix circular definitions?" (Apply)
  • "What causes low concept coverage?" (Analyze)

Best Practices (10% Understand, 40% Apply, 30% Analyze, 15% Evaluate, 5% Create)

Emphasize application and higher-order thinking.

Example Questions:

  • "When should I use a MicroSim vs. a static diagram?" (Evaluate)
  • "How do I balance content depth with cognitive load?" (Apply)
  • "What patterns indicate good chapter structure?" (Analyze)
  • "How could I combine multiple teaching approaches effectively?" (Create)

Advanced Topics (10% Apply, 30% Analyze, 30% Evaluate, 30% Create)

Push toward higher cognitive levels.

Example Questions:

  • "How would you design an adaptive learning system?" (Create)
  • "What are the trade-offs of automated vs. manual content generation?" (Evaluate)
  • "How do different textbook intelligence levels relate to learning outcomes?" (Analyze)

Question Writing Guidelines

For Remember Questions

Do:

  • Ask about specific facts or definitions
  • Use terminology from glossary
  • Keep answers brief (50-150 words)
  • Link to glossary or relevant section

Don't:

  • Ask for explanations or reasoning
  • Require understanding of relationships
  • Make questions too broad

For Understand Questions

Do:

  • Ask "why" or "how"
  • Request explanations or summaries
  • Compare/contrast concepts
  • Include context

Don't:

  • Just ask for definitions (that's Remember)
  • Ask for application (that's Apply)
  • Require evaluation or creation

For Apply Questions

Do:

  • Present specific scenarios
  • Ask "how would you..."
  • Request step-by-step procedures
  • Include practical examples

Don't:

  • Keep questions abstract
  • Just ask for understanding
  • Avoid concrete contexts

For Analyze Questions

Do:

  • Ask about relationships
  • Request identification of patterns
  • Require breaking down into parts
  • Compare underlying structures

Don't:

  • Just list components (that's Remember)
  • Stop at surface-level explanation (that's Understand)
  • Ask for value judgments (that's Evaluate)

For Evaluate Questions

Do:

  • Ask for judgment or decision
  • Present criteria or trade-offs
  • Request recommendations
  • Consider multiple options

Don't:

  • Present only one option
  • Avoid criteria or justification
  • Skip the "why" behind decisions

For Create Questions

Do:

  • Ask for novel solutions
  • Encourage innovation
  • Request design or proposal
  • Allow open-ended responses

Don't:

  • Constrain to single answer
  • Just ask for combining existing elements
  • Make too prescriptive

Common Mistakes and Corrections

Mistake: Mislabeling Bloom's Level

❌ Question: "What is the best way to create a learning graph?"

Incorrectly Labeled: Remember (it's not asking to recall a fact)

Correct Label: Apply (asking how to use knowledge in practice)

Better Phrasing: "How do I create a learning graph for my course?" (clearly Apply)


❌ Question: "List the steps in the ADDIE model"

Incorrectly Labeled: Apply (it's not asking to use the steps)

Correct Label: Remember (asking to recall information)


❌ Question: "Explain why Bloom's Taxonomy is useful"

Incorrectly Labeled: Remember (not just recalling a fact)

Correct Label: Understand (asking for explanation)


Mistake: Questions Too Vague

❌ "How does it work?"

✓ "How does concept dependency validation work in the learning graph generator?"


❌ "What's the difference?"

✓ "What's the difference between a learning graph and a concept taxonomy?"


Mistake: Multiple Questions in One

❌ "What is a MicroSim and how do I create one and when should I use it?"

✓ Split into three questions:

  • "What is a MicroSim?" (Remember)
  • "How do I create a MicroSim?" (Apply)
  • "When should I use a MicroSim vs. a static diagram?" (Evaluate)

Balancing Distribution

Target Overall Distribution

Bloom's Level Target % Question Count (for 100 questions)
Remember 20% 20
Understand 30% 30
Apply 25% 25
Analyze 15% 15
Evaluate 7% 7
Create 3% 3

Acceptable Deviation

Within ±10% for each level is acceptable:

  • Remember: 10-30% ✓
  • Understand: 20-40% ✓
  • Apply: 15-35% ✓
  • Analyze: 5-25% ✓
  • Evaluate: 0-17% ✓
  • Create: 0-13% ✓

Beyond ±10% indicates imbalance and should be corrected.

How to Adjust

Too many Remember/Understand questions?

  • Convert some to Apply by asking "How do I..."
  • Add scenario-based questions
  • Include more troubleshooting questions

Too few Analyze/Evaluate/Create questions?

  • Add comparison questions
  • Include trade-off discussions
  • Ask for recommendations or designs
  • Present complex scenarios

Too many Create questions?

  • Convert some to Evaluate (trade-offs, recommendations)
  • Convert some to Analyze (relationships, patterns)
  • Remember Create should be small % (3% target)

Quality Checklist

For each FAQ question, verify:

  • Correct Bloom's Level: Question matches cognitive action
  • Clear Phrasing: Specific, searchable, unambiguous
  • Appropriate Category: Fits in assigned section
  • Complete Answer: Fully addresses question
  • Appropriate Length: 100-300 words (most questions)
  • Examples Present: 40% of questions have examples
  • Links Present: 60% of questions link to source
  • Terminology Consistent: Uses glossary terms
  • Difficulty Appropriate: Matches category expectations
  • No Duplicates: Unique question not asked elsewhere

Source: SKILL.md on GitHub

2 warnings1mo4 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    The skill uses local scripts and system commands to generate textbook content from project files. It possesses an indirect prompt injection surface due to its data ingestion capabilities combined with file system and subprocess access.

  • Socket1mo

    1 alert: gptSecurity

  • Snyk1mo

    Risk: LOW · No issues

  • Runlayer7mo

    4/5 files flagged

Signed by skilld at 3ccea10. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated yesterday
metadata
{
  "ibook.version": "1.11",
  "ibook.preferred-model": "sonnet"
}

README badge

README badge for dmccreary/claude-skills/chapter-content-generator