Learning Science Expert Skill

A Claude agent skill that provides expertise in learning science, synthesizing research from How People Learn II (2018), practitioner frameworks, and evidence-based instructional design principles.

Overview

This skill enables Claude to act as a learning science consultant, providing:

• Evidence-based recommendations for instructional design
• Pedagogical reviews of courses, curricula, and learning materials
• Guidance on integrating AI into learning contexts
• Research-grounded answers to learning science questions

Installation

Copy the learning-science-expert/ directory to your Claude skills location:

cp -r learning-science-expert/ ~/.claude/skills/

Or reference directly in your Claude configuration.

Usage

The skill activates when users ask about:

• How people learn
• Instructional design principles
• Course or curriculum review
• Educational technology effectiveness
• Motivation, memory, transfer, or other learning science topics

Example Prompts

"Review this course syllabus for alignment with learning science principles"

"What does research say about spaced practice vs. massed practice?"

"How should I design assessments that promote learning, not just measure it?"

"What are the key principles for supporting transfer of learning?"

Architecture

The skill uses a three-layer progressive disclosure architecture:

learning-science-expert/
├── SKILL.md                     # Entry point (<5K tokens)
├── references/
│   ├── index.json               # Topic-to-file mappings
│   ├── core-principles.md       # Synthesized principles
│   ├── domains/                 # 11 topic-specific files
│   │   ├── memory-and-retention.md
│   │   ├── transfer-and-application.md
│   │   ├── expertise-development.md
│   │   ├── motivation-and-identity.md
│   │   ├── culture-and-context.md
│   │   ├── cognitive-load-and-scaffolding.md
│   │   ├── metacognition-and-self-regulation.md
│   │   ├── assessment-for-learning.md
│   │   ├── learning-environments.md
│   │   ├── technology-and-ai.md
│   │   └── lifespan-development.md
│   └── sources/                 # Primary source summaries
│       ├── hpl2/
│       │   ├── overview.md
│       │   └── chapters/
│       └── practitioner-frameworks/
└── assets/
    ├── review-template.md       # Pedagogical review template
    ├── design-checklist.md      # Instructional design checklist
    └── ai-integration-guide.md  # AI in learning contexts

Loading Strategy

1. Always loaded: SKILL.md (~500 words)
2. On demand: Domain files loaded based on query keywords
3. Deep dive: Source files for detailed research questions

Key Frameworks

HPL II Theoretical Shifts

The skill reflects three major shifts from HPL I (2000) to HPL II (2018):

Shift	From	To
Context	Cognition in isolation	Culture as central
Timeframe	Childhood focus	Lifespan perspective
Technology	Passive tool	Active partner (AI)

Nine Principle Categories

1. Prior Knowledge & Transfer
2. Memory & Retention
3. Motivation & Identity
4. Metacognition & Self-Regulation
5. Cognitive Load & Scaffolding
6. Feedback & Assessment
7. Social & Collaborative Learning
8. Cultural Context & Equity
9. Technology & AI Integration

Sources

• How People Learn II (National Academies, 2018)
• The Learning Scientists - Six Strategies for Effective Learning
• RetrievalPractice.org - Evidence-based practice guides
• Deans for Impact - The Science of Learning

Specification Compliance

This skill follows the Agent Skills Specification for portable AI skills:

• Lowercase-hyphenated file naming (except SKILL.md)
• Progressive disclosure architecture
• Token-budgeted files (<5K for SKILL.md, <10K for references)
• JSON index for programmatic access

License

MIT

Contributing

Contributions welcome. Please ensure any additions:

1. Cite peer-reviewed research or established practitioner frameworks
2. Follow the existing file structure and naming conventions
3. Stay within token budgets
4. Update index.json for new topic files