Glossary of Terms
This glossary contains definitions of key concepts used throughout the course "Using Claude Skills to Create Intelligent Textbooks." Each definition follows ISO 11179 metadata registry standards: precise, concise, distinct, non-circular, and free of business rules.
A
Action Verbs for Learning Outcomes
Specific action words that describe observable behaviors students can demonstrate at each level of Bloom's Taxonomy.
Example: The verb "define" indicates Remember level, while "design" indicates Create level in learning outcomes.
add-taxonomy.py Script
A Python program that adds a taxonomy categorization column to a learning graph CSV file.
Example: Running this script on learning-graph.csv adds a TaxonomyID field to categorize each concept.
Advanced Concepts
High-level ideas that require multiple prerequisite concepts to be understood before they can be mastered.
Example: "Capstone: Complete Textbook Project" depends on understanding all 199 prior concepts in the learning graph.
Admonitions in MkDocs
Specially formatted callout boxes in Material for MkDocs that highlight notes, warnings, tips, or important information.
Example: Using !!! note creates a blue box with "Note" header to draw attention to key information.
Allowed Tools in Skills
Specification in skill YAML frontmatter defining which Claude Code tools the skill is permitted to use.
Example: A skill might specify allowed-tools: [Read, Write, Bash] to limit file operations.
Analyze (Cognitive Level 4)
The fourth level of Bloom's Taxonomy where learners break material into parts and determine relationships between components.
Example: Students analyze why a skill fails by examining error logs, file permissions, and workflow dependencies.
analyze-graph.py Script
A Python program that validates learning graph structure, detects circular dependencies, and generates quality metrics reports.
Example: This script checks if your 200-concept graph is a valid DAG and reports statistics like average dependencies per concept.
Anthropic Claude Pro Account
A paid subscription to Claude AI that provides extended usage limits, longer conversation context, and access to Claude Code.
Example: A Claude Pro account allows 4-hour usage windows with higher token limits for generating textbook content.
Apply (Cognitive Level 3)
The third level of Bloom's Taxonomy where learners carry out procedures or use knowledge in specific situations.
Example: Students apply prompt engineering principles to create a new skill for generating chapter quizzes.
Artificial Intelligence
Computer systems designed to perform tasks that typically require human intelligence, such as learning, reasoning, and problem-solving.
Example: Claude AI uses artificial intelligence to understand course descriptions and generate structured learning content.
Assessing Course Descriptions
The process of evaluating a course description for completeness, clarity, and alignment with educational standards.
Example: The course-description-analyzer skill scores descriptions on presence of prerequisites, Bloom's outcomes, and target audience.
Assessing Student Understanding
Methods for evaluating whether learners have mastered concepts through quizzes, exercises, and interactive activities.
Example: Multiple-choice quizzes aligned to Bloom's Taxonomy levels assess student understanding across cognitive domains.
Assessing Student Understanding Through Quiz Analytics
Evaluation of learner comprehension by analyzing quiz performance patterns across cognitive levels and concept areas to identify knowledge gaps.
Example: Analyzing quiz results shows students score well on Remember questions but struggle with Analyze questions, indicating need for more worked examples.
Atomic Concepts
Single, indivisible ideas that cannot be meaningfully broken into smaller learning components.
Example: "Variable" is atomic, while "Variable Declaration and Assignment" could be split into two atomic concepts.
Average Dependencies Per Concept
The mean number of prerequisite relationships each concept has in a learning graph.
Example: A well-structured graph typically has 2-4 average dependencies per concept, avoiding both isolation and over-complexity.
Avoiding Over-Representation
The practice of ensuring no single taxonomy category contains too many concepts, maintaining balanced distribution.
Example: If 60% of concepts are in the BASIC category, the graph may need rebalancing across foundational and advanced levels.
B
Bash
A Unix shell and command language used for executing system commands, running scripts, and automating tasks.
Example: Running ./install-claude-skills.sh in Bash creates symlinks to make skills available globally.
Bloom's 2001 Revision
An updated framework of cognitive learning objectives that replaced "Knowledge" with "Remember" and "Synthesis" with "Create."
Example: The 2001 revision emphasizes active learning verbs and clearer distinctions between cognitive levels.
Bloom's Taxonomy
A hierarchical framework of six cognitive levels used to classify educational learning objectives from simple recall to complex creation.
Example: Learning outcomes progress from Remember (defining terms) through Create (designing complete textbooks).
Bloom's Taxonomy in Quizzes
The practice of distributing quiz questions across all six cognitive levels to assess comprehensive understanding.
Example: A quiz includes 20% Remember questions, 20% Understand, 15% Apply, 20% Analyze, 15% Evaluate, and 10% Create.
C
Capstone: Complete Textbook Project
A culminating project where learners design and implement an entire intelligent textbook from course description through deployment.
Example: The capstone requires creating a learning graph, generating content, building MicroSims, and publishing to GitHub Pages.
Category Distribution
The spread of concepts across different taxonomy categories in a learning graph.
Example: A balanced distribution might be 15% foundational, 35% basic, 30% intermediate, 15% advanced, and 5% integration concepts.
Chapter Concept Lists
Enumerated sets of specific concepts that will be covered within a particular chapter of a textbook.
Example: Chapter 3's concept list includes concepts 45-67 from the learning graph, respecting dependency order.
Chapter Index Files
Markdown files named index.md that serve as the main content page for each chapter in a MkDocs textbook.
Example: /docs/chapters/03/index.md contains the title, summary, concept list, and full content for Chapter 3.
Chapter Structure
The organizational framework defining how textbook content is divided into major sections with logical progression.
Example: A 200-concept course might be organized into 12 chapters with 15-20 concepts per chapter.
Chapter Structure and Token Budgeting
The dual consideration of logical content organization and AI token limit management when designing textbook chapters.
Example: Balancing pedagogical needs (15-20 concepts per chapter) with practical constraints (staying under Claude's token limits per generation).
Circular Dependency Detection
The process of identifying invalid prerequisite loops where concept A depends on B, which depends on C, which depends on A.
Example: If "Variables" requires "Functions" which requires "Variables," the analyze-graph.py script reports a circular dependency error.
Claude AI
An artificial intelligence assistant created by Anthropic that uses large language models to understand and generate human-like text.
Example: Claude AI can read a course description and generate a complete 200-concept learning graph with dependencies.
Claude Code Interface
The command-line tool that enables users to interact with Claude AI for software development and content creation tasks.
Example: Running claude in the terminal launches an interactive session where you can invoke skills and execute commands.
Claude Command
A user-defined operation in Claude Code that expands a slash command into a full prompt for common workflows.
Example: The /skills command expands to list all available Claude skills in the current project.
Claude Pro Limitations
Usage restrictions on Claude Pro accounts including token limits per message and 4-hour usage windows.
Example: Claude Pro allows higher token limits than free accounts but still requires managing usage within 4-hour windows.
Claude Skill
An autonomous agent defined by a SKILL.md file that automates specific tasks in the Claude Code environment.
Example: The learning-graph-generator skill automates creating a 200-concept dependency graph from a course description.
Claude Token Limits
Maximum number of tokens (roughly word pieces) that can be processed in a single Claude conversation or message.
Example: Managing token limits requires breaking large content generation into multiple skill invocations.
Color Coding in Visualizations
Using distinct colors to represent different categories, groups, or properties in graphical displays.
Example: Learning graph visualizations use color to distinguish foundational concepts (orange) from advanced concepts (purple).
Command-Line Interface Basics
Fundamental concepts and operations for interacting with computers through text-based terminal commands.
Example: Basic CLI skills include navigating directories with cd, listing files with ls, and running scripts.
Command Definition Files
Markdown files that specify slash command behaviors, stored in the .claude/commands/ directory.
Example: The file .claude/commands/skills.md defines what happens when you type /skills.
Common Student Questions
Frequently asked queries that learners typically have about course content, processes, or concepts.
Example: "How do I install a skill globally versus project-specific?" is a common question addressed in the FAQ.
Concept Categorization
The process of organizing concepts into groups based on difficulty level, subject area, or other distinguishing characteristics.
Example: Categorizing concepts as foundational, basic, intermediate, or advanced helps structure curriculum progression.
Concept Dependencies
Prerequisite relationships where understanding one concept requires prior mastery of other specific concepts.
Example: Understanding "Dependency Edges in Learning Graphs" depends on first understanding "Learning Graph."
Concept Enumeration Process
The systematic method of identifying and listing all atomic concepts that comprise a course or subject area.
Example: Reading the course description and generating exactly 200 distinct, atomic concepts covering all main topics.
Concept Granularity
The level of detail or specificity at which ideas are broken down into individual learning components.
Example: "Git Commands" has low granularity, while separate concepts for "Git Add Command," "Git Commit Command" has higher granularity.
Concept Label Requirements
Specifications that concept names must follow, including Title Case formatting and maximum character length constraints.
Example: Concept labels must be in Title Case and not exceed 32 characters to ensure readability in visualizations.
ConceptID Field
A unique numeric identifier assigned to each concept in a learning graph CSV file.
Example: The ConceptID field contains integers from 1 to 200, providing a stable reference for each concept.
ConceptLabel Field
The human-readable name of a concept in a learning graph CSV file, following Title Case and length conventions.
Example: The ConceptLabel field might contain "Learning Graph" or "Directed Acyclic Graph (DAG)."
Concept Nodes in Learning Graphs
Individual concepts represented as vertices in a directed graph structure showing learning relationships.
Example: In a visualization, each concept appears as a labeled circle (node) with arrows (edges) pointing to dependent concepts.
Concise Definitions
Brief explanations that convey essential meaning using minimal words, typically 20-50 words for glossary entries.
Example: "A directed graph of concepts" is more concise than "A specialized type of graph structure that shows relationships."
Content Generation Process
The systematic workflow for creating textbook chapters, sections, and supporting materials using AI assistance.
Example: The chapter-content-generator skill reads concept lists and generates comprehensive content with examples and exercises.
Course Description
A comprehensive document defining a course's title, audience, prerequisites, topics, exclusions, and Bloom's Taxonomy-aligned outcomes.
Example: A complete course description enables the learning-graph-generator skill to identify relevant concepts and dependencies.
Course Description Quality Score
A numeric assessment (1-100) evaluating how well a course description meets completeness and clarity standards.
Example: A score of 95 indicates all required sections are present with clear, measurable learning outcomes.
Course Prerequisites
Knowledge, skills, or experiences that learners must possess before beginning a course.
Example: Prerequisites for this course include basic programming understanding and access to Claude Pro.
Create (Cognitive Level 6)
The highest level of Bloom's Taxonomy where learners put elements together to form coherent, original works.
Example: Students create new Claude skills from scratch, designing workflows and writing skill definition files.
Creator Metadata Field
Dublin Core element identifying the person, organization, or entity responsible for creating a resource.
Example: The creator field in metadata.json might contain "Dan McCreary" or your organization name.
csv-to-json.py Script
A Python program that converts learning graph CSV files into vis-network JSON format for visualization.
Example: Running this script transforms learning-graph.csv into learning-graph.json with nodes and edges arrays.
CSV File Format for Graphs
A structured text format using comma-separated values to store learning graph data with headers and rows.
Example: Graph CSVs contain columns: ConceptID, ConceptLabel, Dependencies, TaxonomyID.
D
DAG Validation
The process of verifying that a learning graph forms a valid directed acyclic graph with no circular dependencies.
Example: The analyze-graph.py script performs DAG validation and reports any cycles that would prevent topological sorting.
Date Metadata Field
Dublin Core element recording when a resource was created, modified, or published.
Example: The date field captures "2025-11-08" as the generation date for the learning graph.
Definitions Without Business Rules
ISO 11179 principle that glossary definitions should describe concepts without prescribing processes or requirements.
Example: Define "Prerequisites" as relationships between concepts, not "Students must complete prerequisites before advancing."
Dependency Edges in Learning Graphs
Directed arrows connecting prerequisite concepts to dependent concepts in a graph structure.
Example: An edge from "Variables" to "Functions" indicates that understanding variables is prerequisite for understanding functions.
Dependency Mapping Process
The systematic method of identifying and recording prerequisite relationships between all concepts in a learning graph.
Example: For each of 200 concepts, determine which prior concepts must be understood first and record them in the Dependencies field.
Dependencies Field
A CSV column containing pipe-delimited ConceptIDs representing prerequisite concepts that must be learned first.
Example: A Dependencies field of "1|5|12" means concepts 1, 5, and 12 are prerequisites for the current concept.
Description Metadata Field
Dublin Core element providing a textual explanation of a resource's content and purpose.
Example: The description field summarizes what the learning graph covers and its educational objectives.
Difference Between Skills & Commands
Skills are autonomous workflows defined in SKILL.md files, while commands are simple prompt expansions for common tasks.
Example: A skill generates entire chapters; a command might just list available skills or clear the conversation history.
Directed Acyclic Graph (DAG)
A graph structure with directed edges and no cycles, where you cannot traverse from any node back to itself.
Example: Learning graphs must be DAGs to ensure a valid learning sequence exists without circular prerequisites.
Directory Navigation
The process of moving between folders in a file system using command-line or graphical interfaces.
Example: Use cd docs/chapters to navigate into the chapters directory from the project root.
Disconnected Subgraphs
Separate groups of concepts in a learning graph that have no dependency connections between groups.
Example: If web development concepts and database concepts form separate clusters with no links, they are disconnected subgraphs.
Distinct Definitions
ISO 11179 principle ensuring each glossary entry is clearly distinguishable from related terms.
Example: "Learning Graph" and "Concept Dependency" have distinct definitions focusing on different aspects of concept relationships.
Dublin Core Metadata
A standardized set of 15 metadata elements for describing digital resources, including title, creator, date, and format.
Example: MicroSim metadata.json files use Dublin Core fields to document simulation properties.
E
Edges Section in JSON
The array in vis-network JSON format containing objects that define directed connections between concept nodes.
Example: {"from": 1, "to": 2} in the edges array represents a dependency from concept 1 to concept 2.
Educational Content Prompts
Carefully designed instructions to AI systems specifying how to generate pedagogically sound learning materials.
Example: A prompt might request "Generate 5 worked examples demonstrating this concept at the Apply level of Bloom's Taxonomy."
Educational Simulation Design
The process of creating interactive visual models that demonstrate concepts through user manipulation and observation.
Example: A MicroSim showing bubble sort lets learners control array size and see step-by-step swaps in real-time.
Error Analysis in Skills
The systematic examination of skill execution failures to identify root causes and improvement opportunities.
Example: Analyzing error messages reveals that a skill failed because it expected a file in /docs/learning-graph/ that didn't exist.
Evaluate (Cognitive Level 5)
The fifth level of Bloom's Taxonomy where learners make judgments based on criteria and standards.
Example: Students evaluate the quality of a generated glossary using ISO 11179 compliance as evaluation criteria.
F
FAQ
A document containing Frequently Asked Questions and their answers to help learners find common information quickly.
Example: The FAQ addresses questions like "How do I install skills globally?" and "What's the difference between skills and commands?"
FAQ from Course Content
Questions and answers derived from analyzing course materials, learning graphs, and glossary terms.
Example: Generated FAQs explain technical terms from the glossary in simpler language with additional context.
FAQ Generation Process
The systematic workflow for creating comprehensive question-answer pairs from course content and anticipated learner needs.
Example: The faq-generator skill scans course content, identifies potential confusion points, and creates clear Q&A pairs.
File Access Permissions
Security settings that control which users and processes can read, write, or execute specific files.
Example: Script execution requires permission: chmod +x install-claude-skills.sh makes the file executable.
File Creation and Editing
The process of making new files or modifying existing files using text editors or command-line tools.
Example: Use touch glossary.md to create a new file, then edit it with VS Code or vim.
Five Levels of Textbook Intelligence
A framework categorizing educational materials from static text (Level 1) to AI-personalized learning (Level 5).
Example: Traditional PDFs are Level 1, while textbooks with adaptive quizzes and learning paths are Level 4.
Font Colors for Readability
Strategic color choices for text that ensure sufficient contrast and accessibility across different backgrounds.
Example: Light text on dark nodes requires careful color selection to maintain readability in graph visualizations.
Format Metadata Field
Dublin Core element specifying the file format or media type of a resource.
Example: The format field might indicate "text/html" for MicroSims or "application/json" for learning graphs.
Foundational Concepts
Fundamental ideas with no prerequisites that serve as building blocks for more advanced understanding.
Example: "Artificial Intelligence" and "Claude AI" are foundational concepts that don't depend on other course concepts.
4-Hour Usage Windows
Time-based limits on Claude Pro usage where token allowances reset after four hours of activity.
Example: Planning to generate multiple chapters within a single 4-hour window maximizes efficiency.
G
Generating 200 Concepts
The process of systematically enumerating exactly 200 atomic concepts that comprehensively cover a course's scope.
Example: The learning-graph-generator skill analyzes the course description and produces a numbered list of 200 concepts.
Git
A distributed version control system for tracking changes in source code and collaborating on software projects.
Example: Git commands like git add, git commit, and git push manage textbook content versions.
Git Add Command
A Git operation that stages modified or new files for inclusion in the next commit.
Example: git add docs/glossary.md stages the glossary file for committing to the repository.
Git Commit Command
A Git operation that saves staged changes to the local repository with a descriptive message.
Example: git commit -m "Add glossary with 200 ISO 11179-compliant definitions" records the glossary addition.
Git Push Command
A Git operation that uploads local commits to a remote repository like GitHub.
Example: git push origin main sends your latest commits to the main branch on GitHub.
Git Repository Structure
The organization of files, directories, and Git metadata within a version-controlled project.
Example: A textbook repo includes /docs, /skills, .git hidden folder, and configuration files like mkdocs.yml.
Git Status Command
A Git operation that displays the current state of the working directory and staging area.
Example: git status shows which files are modified, staged, or untracked.
GitHub Integration
The connection between local Git repositories and GitHub's cloud-based platform for hosting and collaboration.
Example: Linking a local textbook repository to GitHub enables deployment to GitHub Pages.
GitHub Pages Deployment
The process of publishing static website content to GitHub's free web hosting service.
Example: Running mkdocs gh-deploy builds the site and pushes it to the gh-pages branch for public access.
Glossary
An alphabetically organized collection of terms and their definitions specific to a subject area or course.
Example: This glossary defines 200 concepts related to creating intelligent textbooks with Claude Skills.
Glossary Generation Process
The systematic workflow for creating ISO 11179-compliant definitions from a concept list with quality validation.
Example: The glossary-generator skill reads concept lists, generates definitions, adds examples, and produces quality reports.
Groups Section in JSON
The array in vis-network JSON format defining visual properties for each taxonomy category.
Example: Groups specify colors like orange for foundational concepts and purple for advanced concepts.
I
Iframe Embedding
The technique of inserting one HTML document inside another using the <iframe> tag.
Example: MicroSim documentation embeds main.html simulations in index.md pages using iframe tags.
Improving Skill Quality
The iterative process of refining skill workflows, error handling, and output quality through testing and feedback.
Example: Adding validation checks to skills ensures they fail gracefully when required input files are missing.
Indegree Analysis
Examining how many prerequisite concepts point to each concept in a learning graph.
Example: High indegree concepts require many prerequisites and are typically advanced topics.
index.md for MicroSim Docs
A markdown documentation file that describes a MicroSim's purpose, usage, and embeds the interactive simulation.
Example: /docs/sims/bubble-sort/index.md explains the simulation and embeds main.html in an iframe.
Installing a Claude Skill
The process of making a skill available for use in Claude Code by placing it in the .claude/skills/ directory.
Example: Running ./scripts/install-claude-skills.sh creates symlinks from project skills to global skills directory.
Installing Claude Commands
The process of placing command definition files in .claude/commands/ to enable slash command usage.
Example: Copying skills.md to .claude/commands/ enables the /skills command.
Installing Python Packages
The process of adding third-party Python libraries to your environment using package management tools.
Example: pip install mkdocs-material installs the Material theme for MkDocs documentation.
Installing Skills Globally
Placing skills in ~/.claude/skills/ to make them available across all Claude Code projects.
Example: Global skills can be invoked from any project directory without per-project installation.
Intelligent Textbook
An educational resource enhanced with interactive elements, adaptive content, or AI-powered features beyond static text.
Example: An intelligent textbook includes learning graphs, interactive MicroSims, and automatically generated quizzes.
Interactive Controls (Buttons)
User interface elements in MicroSims that trigger actions or state changes when clicked.
Example: A "Reset" button returns the simulation to initial conditions, while "Next Step" advances one iteration.
Interactive Controls (Sliders)
User interface elements in MicroSims that allow continuous value adjustment through dragging or clicking.
Example: A slider controls animation speed from 1 to 100, letting learners observe processes at different rates.
Interactive Elements
Components in educational materials that respond to user input and provide dynamic feedback.
Example: Interactive elements include clickable diagrams, adjustable parameters in simulations, and self-grading quizzes.
Interactive Simulations
Computer programs that model real-world or abstract processes, allowing learners to manipulate variables and observe outcomes.
Example: A physics simulation lets students adjust mass and velocity to see how momentum changes.
Invoking Skills with Slash Commands
The method of executing Claude skills by typing /skill [skill-name] in the Claude Code interface.
Example: Typing the skill name launches the skill to create a 200-concept dependency graph.
ISO 11179 Standards
International metadata registry standards defining principles for creating precise, concise, distinct, non-circular definitions.
Example: ISO 11179 requires glossary definitions avoid business rules like "must complete before advancing."
Iterative Prompt Refinement
The process of progressively improving AI prompts through testing, evaluation, and modification based on results.
Example: Refining a content generation prompt by adding "include 3 worked examples" after initial results lacked examples.
J
JSON Schema for Learning Graphs
A formal specification defining the required structure and data types for learning graph JSON files.
Example: The schema requires "nodes" and "edges" arrays with specific properties like "id," "label," "from," and "to."
L
Large Language Models Overview
An introduction to AI systems trained on vast text corpora to understand and generate human-like language.
Example: Claude, GPT, and other large language models can generate educational content from structured prompts.
Learning Graph
A directed graph of concepts showing prerequisite relationships that guide the optimal sequence for learning material.
Example: A learning graph for programming shows that "Variables" must be understood before "Functions," which precedes "Recursion."
Learning Graph Quality Score
A numeric assessment (1-100) evaluating graph structure quality based on connectivity, balance, and DAG validity.
Example: A score of 89 indicates excellent structure with balanced dependencies and no circular references.
Learning Outcomes
Specific, measurable statements describing what learners will be able to do after completing a course or module.
Example: "Students will be able to create a learning graph with 200 concepts and validate it has no circular dependencies."
Learning Pathways
Recommended sequences through course material that respect prerequisite dependencies and learner goals.
Example: A pathway for beginners starts with foundational concepts, while advanced learners might skip to intermediate concepts.
Level 1: Static Content
Textbooks containing only fixed text and images with no interactive or dynamic features.
Example: A PDF textbook or printed book represents Level 1 intelligence.
Level 2: Hyperlinked Navigation
Textbooks with clickable links enabling non-linear exploration and cross-referencing between sections.
Example: MkDocs-generated websites provide Level 2 intelligence with navigation menus and internal links.
Level 3: Interactive Elements
Textbooks incorporating user-controlled components like simulations, quizzes, and dynamic visualizations.
Example: Adding p5.js MicroSims and self-grading quizzes elevates a textbook to Level 3.
Level 4: Adaptive Content
Textbooks that adjust material presentation based on learner performance, preferences, or knowledge gaps.
Example: A Level 4 textbook recommends review material when quiz scores indicate weak understanding of prerequisites.
Level 5: AI Personalization
Textbooks using artificial intelligence to create customized learning experiences for individual learners.
Example: A Level 5 textbook generates practice problems tailored to each student's specific misconceptions.
License Metadata Field
Dublin Core element specifying usage rights, restrictions, and permissions for a resource.
Example: The license field might indicate "CC-BY-4.0" for Creative Commons Attribution license.
Linear Chain Detection
Identifying sequences in a learning graph where concepts form a single dependency line without branching.
Example: A linear chain like "A→B→C→D" lacks the richness of interconnected prerequisite relationships.
Listing Available Skills
The process of displaying all Claude skills accessible in the current project or globally.
Example: Running /skills or ./scripts/list-skills.sh shows all skills with their descriptions.
M
main.html in MicroSims
The standalone HTML file containing complete p5.js simulation code that can run independently in a browser.
Example: /docs/sims/sorting/main.html contains the full bubble sort visualization with embedded JavaScript.
Main Topics Covered
A comprehensive list of subject areas and themes included within a course's scope.
Example: Main topics include Claude Skills architecture, learning graphs, MkDocs, and Bloom's Taxonomy application.
Markdown Formatting Basics
Fundamental syntax for creating formatted text documents using plain text with special character conventions.
Example: Use # for headers, **text** for bold, *text* for italic, and - for bullet lists.
Maximum Character Length
The constraint that concept labels should not exceed 32 characters to ensure readability in visualizations and tables.
Example: "Directed Acyclic Graph (DAG)" is exactly 29 characters, fitting the maximum length requirement.
Maximum Dependency Chain Length
The longest path through prerequisite relationships from a foundational concept to a terminal concept.
Example: A chain length of 15 means some concepts require understanding 14 prerequisite concepts in sequence.
Metadata Section in JSON
The top-level object in vis-network JSON containing Dublin Core fields describing the learning graph resource.
Example: Metadata includes title, creator, description, date, version, format, and license information.
MicroSim
A focused, interactive p5.js simulation demonstrating a single educational concept through visual manipulation and observation.
Example: A binary search MicroSim shows how the algorithm eliminates half the search space with each comparison.
MicroSim Directory Structure
The standardized organization of files within a simulation folder: main.html, index.md, and metadata.json.
Example: /docs/sims/bubble-sort/ contains these three files for the bubble sort visualization.
MicroSim Metadata
Dublin Core fields stored in metadata.json describing a simulation's title, creator, description, and educational purpose.
Example: Metadata records the concept being taught, target audience, and date created.
MicroSim Metadata Standards
The Dublin Core specification requirements for documenting simulations including required fields (title, creator, date) and recommended fields (description, version, license).
Example: MicroSim Metadata Standards ensure all simulations include consistent documentation for discovery, attribution, and educational context.
MkDocs
A static site generator that creates documentation websites from markdown files, designed for project documentation.
Example: Running mkdocs build converts markdown files in /docs into HTML pages with navigation.
MkDocs Configuration File
The mkdocs.yml file defining site settings, theme, navigation structure, and plugin configurations.
Example: The configuration file specifies the Material theme, navigation menu, and site name.
MkDocs Material Theme
A popular responsive theme for MkDocs providing modern design, search, and extensive customization options.
Example: Material theme enables features like tabs, admonitions, code highlighting, and dark mode.
Multiple-Choice Questions
Quiz items presenting a question with several answer options where learners select the correct response.
Example: "Which command stages files for commit? A) git push B) git add C) git status D) git clone"
N
Navigation Structure in MkDocs
The hierarchical menu organization defined in mkdocs.yml that controls how pages appear in site navigation.
Example: The nav section defines chapters, subsections, and page ordering in the left sidebar menu.
Nodes Section in JSON
The array in vis-network JSON format containing objects representing individual concepts with id, label, and group properties.
Example: {"id": 5, "label": "Claude Code Interface", "group": "FOUND"} defines one concept node.
Non-Circular Definitions
ISO 11179 principle requiring definitions avoid referencing the term being defined or creating circular chains.
Example: Don't define "Learning Graph" using "graph for learning"; instead describe its structure and purpose.
O
Optimizing Claude Usage
Strategies for maximizing productivity within Claude's token limits and usage windows through efficient prompting and batching.
Example: Generate multiple chapters in one session rather than invoking skills separately for each chapter.
Orphaned Nodes
Concepts in a learning graph that no other concepts depend on, suggesting they may be too specific or misplaced.
Example: If concept 150 has no concepts listing it as a prerequisite, it's orphaned and may need reevaluation.
Outdegree Analysis
Examining how many other concepts depend on each concept as a prerequisite in a learning graph.
Example: High outdegree indicates fundamental concepts that enable understanding of many subsequent topics.
P
p5.js JavaScript Library
An open-source JavaScript library for creating interactive graphics and animations, especially suited for educational visualizations.
Example: p5.js provides simple functions like circle() and line() for drawing, plus setup() and draw() for animation loops.
Permission Management
The system for controlling access rights to files, directories, and commands in operating systems.
Example: Skills require read permissions on course files and write permissions on output directories.
pip Package Management
Python's standard tool for installing, upgrading, and managing third-party libraries and dependencies.
Example: pip install mkdocs downloads and installs MkDocs and its dependencies.
Pipe-Delimited Dependencies
A format for recording multiple prerequisite concept IDs separated by vertical bar characters in CSV files.
Example: The Dependencies field "1|5|12" indicates concepts 1, 5, and 12 are all prerequisites.
Practice Exercises
Learning activities where students apply concepts to solve problems, reinforcing understanding through active practice.
Example: After learning about skills, students practice creating a custom skill for their own use case.
Precise Definitions
ISO 11179 principle requiring glossary entries accurately capture specific meanings without ambiguity.
Example: Define "DAG" as "directed acyclic graph" with structural properties, not vaguely as "a type of graph."
Prerequisite Concepts
Ideas that must be understood before a learner can successfully grasp dependent advanced concepts.
Example: "Variables" is a prerequisite concept for understanding "Functions" in programming.
Prerequisite Relationships
Dependency connections indicating that understanding one concept requires prior mastery of other specific concepts.
Example: The relationship between "Git" and "Git Add Command" shows the general concept must precede the specific command.
Project-Specific Skills
Claude skills installed in a project's .claude/skills/ directory, available only within that project.
Example: A custom textbook-generation skill stored locally serves project-specific needs without global installation.
Prompt Design Principles
Guidelines for creating effective AI instructions including clarity, specificity, context provision, and example usage.
Example: Good prompts specify output format, provide context, include examples, and define success criteria.
Prompt Engineering
The practice of crafting precise instructions to AI systems to elicit desired outputs and behaviors.
Example: Prompt engineering transforms "write about graphs" into "generate 200 atomic concepts with dependencies for graph theory."
Python
A high-level programming language widely used for scripting, data processing, and automation tasks.
Example: Python scripts like analyze-graph.py process learning graph data and generate quality reports.
Python Scripts for Processing
Automated programs written in Python to transform, analyze, or validate data in the textbook creation workflow.
Example: Scripts convert CSV to JSON, add taxonomy categories, and analyze graph structure.
Python Scripts in Skills
Supporting Python programs included in skill directories that perform specific data processing tasks.
Example: The learning-graph-generator skill includes four Python scripts for validation and transformation.
Q
Quality Metrics for Graphs
Quantitative measures used to assess learning graph structure, including connectivity, balance, and DAG validity.
Example: Quality metrics include average dependencies, chain length, orphaned nodes, and category distribution percentages.
Quiz
A set of questions designed to assess learner understanding of concepts, typically with immediate feedback.
Example: Chapter quizzes contain 10-15 multiple-choice questions distributed across Bloom's Taxonomy levels.
Quiz Alignment with Concepts
The practice of ensuring each quiz question specifically tests understanding of identified learning graph concepts.
Example: Question 5 tests concept 42 ("Directed Acyclic Graph") by asking students to identify invalid graph structures.
Quiz Distribution Across Levels
Spreading quiz questions proportionally across all six Bloom's Taxonomy cognitive levels for comprehensive assessment.
Example: A balanced quiz has 15% Remember, 20% Understand, 20% Apply, 20% Analyze, 15% Evaluate, 10% Create questions.
R
Reading Level Appropriateness
Ensuring textbook language complexity matches the target audience's comprehension abilities.
Example: Professional development content uses more technical vocabulary than high school materials.
Reading Level Categories
Standard classifications of target audiences based on educational development stage: junior-high, senior-high, college, and graduate levels.
Example: Reading Level Categories determine vocabulary complexity, sentence structure, and conceptual depth for generated textbook content.
Reference Documentation in Skills
Supporting files in skill directories providing examples, specifications, or guidance for skill execution.
Example: The learning-graph-generator includes sample CSVs and JSON schemas as reference documentation.
Remember (Cognitive Level 1)
The first level of Bloom's Taxonomy where learners retrieve, recognize, and recall relevant knowledge from memory.
Example: Students remember the six levels of Bloom's Taxonomy: Remember, Understand, Apply, Analyze, Evaluate, Create.
S
Script Execution Permissions
File system settings that determine whether a script file can be run as a program.
Example: chmod +x script.sh grants execution permission, allowing ./script.sh to run.
Section Organization
The structure of how content within chapters is divided into logical subsections with clear headings.
Example: Sections might include Introduction, Core Concepts, Examples, Practice Exercises, and Summary.
Security in Skill Execution
Safeguards ensuring skills only access authorized files and don't perform unintended or harmful operations.
Example: Skills should validate input file existence before processing to prevent unintended file creation.
Seeded Randomness
Using a fixed starting value for random number generation to produce reproducible results across simulation runs.
Example: randomSeed(42) in p5.js ensures the same "random" pattern appears each time for consistent demonstrations.
Self-Dependency Checking
Validation that no concept lists itself as its own prerequisite in a learning graph.
Example: The analyze-graph.py script reports an error if concept 25 includes 25 in its Dependencies field.
Shell Scripts
Text files containing sequences of command-line instructions that can be executed as automated programs.
Example: install-claude-skills.sh is a shell script that creates multiple symlinks with one command.
Skill Definition File Structure
The required format for SKILL.md files including YAML frontmatter followed by markdown workflow instructions.
Example: Frontmatter specifies name, description, and license; content describes step-by-step execution workflow.
Skill Directory Structure
The standardized organization of a skill folder containing SKILL.md, supporting scripts, templates, and reference files.
Example: learning-graph-generator/ contains SKILL.md plus four Python scripts for graph processing.
Skill Distribution Methods
Approaches for sharing Claude skills including global installation, project-specific placement, or package repositories.
Example: Distribution methods include symlinks to global directory, copying to project folders, or Git repositories.
Skill Execution Context
The environment information available when a skill runs, including working directory, available files, and user permissions.
Example: Skills execute from the project root directory with access to files the user can read or write.
Skill License Information
The legal terms specifying how a skill can be used, modified, and distributed by others.
Example: Most skills use Apache-2.0 license allowing free use with attribution.
Skill Name and Description
The identifying label and brief summary in YAML frontmatter explaining what a skill does.
Example: name: glossary-generator and description: Generate ISO 11179-compliant glossaries from concept lists.
Skill Packaging Best Practices
Guidelines for organizing, documenting, and distributing skills to ensure usability and maintainability.
Example: Include README files, example inputs/outputs, clear dependencies, and comprehensive skill documentation.
Skill Testing and Debugging
The process of validating skill functionality, identifying errors, and iteratively improving skill reliability.
Example: Test skills with missing input files, malformed data, and edge cases to ensure robust error handling.
Skill Workflow Instructions
Detailed step-by-step procedures in SKILL.md that guide Claude through executing the skill's tasks.
Example: Workflow instructions specify "Step 1: Read concept list. Step 2: Validate quality. Step 3: Generate definitions."
Supporting Assets in Skills
Additional files in skill directories that assist execution, including scripts, templates, examples, and schemas.
Example: Supporting assets might include Python scripts, sample CSVs, JSON schemas, or reference documentation.
Symlink Creation
Making symbolic links that reference files in other locations, allowing access from multiple paths without duplication.
Example: ln -s ~/claude-skills/skills ~/.claude/skills creates a symlink for global skill access.
T
Target Audience Definition
A clear specification of who the course is designed for, including their background and learning goals.
Example: "Professional development for educators with basic programming skills who want to create AI-assisted textbooks."
Taxonomy
A hierarchical classification system organizing concepts into categories based on shared characteristics or difficulty levels.
Example: Concepts are categorized as FOUND (foundational), BASIC, INTER (intermediate), ADVAN (advanced), or INTEG (integration).
Taxonomy Categories
Distinct groupings within a taxonomy system used to organize concepts by type, difficulty, or subject area.
Example: Common categories include foundational, basic, intermediate, advanced, and integration concepts.
taxonomy-distribution.py Script
A Python program that analyzes and reports the distribution of concepts across taxonomy categories.
Example: Running this script produces a report showing 15% foundational, 40% basic, 30% intermediate, 15% advanced concepts.
TaxonomyID Abbreviations
Short alphanumeric codes representing concept categories in learning graphs, typically 3-5 characters.
Example: FOUND for foundational, BASIC for basic concepts, ADVAN for advanced concepts.
TaxonomyID Field in CSV
A column in learning graph CSV files containing category abbreviations for each concept.
Example: The TaxonomyID field might contain "FOUND" for concept 1 and "ADVAN" for concept 200.
Template Files in Skills
Reusable file structures that skills populate with generated content to ensure consistent formatting.
Example: A chapter template defines standard sections: Introduction, Concepts, Examples, Exercises, Summary.
Terminal Commands
Text-based instructions entered in a command-line interface to execute programs, manage files, or configure systems.
Example: Terminal commands include cd, ls, mkdir, python, and git for various operations.
Terminal in VS Code
An integrated command-line interface within Visual Studio Code for running commands without leaving the editor.
Example: Open the VS Code terminal with Ctrl+to runmkdocs serve` while editing content.
The Role of FAQs in Intelligent Textbooks
The function of Frequently Asked Questions as supplementary resources providing quick answers to common learner queries and clarifying potentially confusing concepts.
Example: FAQs bridge gaps between formal content and student questions, offering conversational explanations that complement technical definitions in the glossary.
Title Case Convention
Capitalization style where the first letter of each major word is uppercase, used for concept labels.
Example: "Learning Graph Generator" follows Title Case, while "learning graph generator" does not.
Title Metadata Field
Dublin Core element providing the formal name or title of a resource.
Example: The title field might contain "Learning Graph for Claude Skills Course."
Token Management Strategies
Techniques for working within Claude's token limits including content chunking, selective context, and multi-turn workflows.
Example: Generate one chapter at a time rather than requesting all chapters in a single prompt.
Topics Excluded from Course
Subjects explicitly identified as out of scope to clarify course boundaries and manage learner expectations.
Example: This course excludes advanced machine learning theory and general web development beyond MkDocs.
U
Understand (Cognitive Level 2)
The second level of Bloom's Taxonomy where learners construct meaning from instructional messages and explanations.
Example: Students understand how learning graphs guide concept sequencing by explaining prerequisite relationships.
V
Version Control Basics
Fundamental concepts of tracking changes to files over time, including commits, branches, and repositories.
Example: Version control lets you see who changed what, when, and why, plus revert to earlier versions if needed.
Version Metadata Field
Dublin Core element recording the iteration or release number of a resource.
Example: The version field tracks "1.0" for initial release or "2.3" after multiple revisions.
vis-network JSON Format
A specific JSON structure used by the vis-network visualization library containing nodes, edges, groups, and metadata.
Example: The format requires separate arrays for nodes (with id, label, group) and edges (with from, to).
Visual Studio Code
A popular open-source code editor with integrated terminal, Git support, and extensions for various development tasks.
Example: VS Code provides markdown preview, making it ideal for writing and editing textbook content.
VS Code for Content Development
Using Visual Studio Code's features specifically for creating and managing educational textbook content.
Example: VS Code extensions for markdown linting, spell-checking, and MkDocs preview enhance content development.
W
Worked Examples in Content
Step-by-step demonstrations showing how to solve problems or apply concepts with detailed explanations.
Example: A worked example shows creating a 10-concept learning graph from scratch, explaining each dependency decision.
Y
YAML Frontmatter in Skills
Metadata section at the beginning of SKILL.md files enclosed in --- delimiters containing key-value pairs.
Example: Frontmatter includes name: glossary-generator, description: Generate glossaries, and license: Apache-2.0.