This Claude Code skill addresses the risk of AI coding tools diminishing developer learning by introducing deliberate skill-building exercises. Targeting developers using agentic AI coding, it offers a "dynamic textbook" approach to integrate evidence-based learning techniques, fostering deeper expertise and long-term retention rather than just project completion.

How It Works

The skill operates by offering optional 10-15 minute learning exercises after significant architectural work, such as creating new files, schema changes, or refactors. It employs techniques like prediction, generation, retrieval practice, and spaced repetition, intentionally pausing for user input to encourage active mental effort. This approach directly counteracts common AI coding pitfalls like the generation effect and fluency illusion by reintroducing deliberate learning pauses and metacognitive reflection.

Quick Start & Requirements

• Installation: Add the marketplace (/plugin marketplace add https://github.com/DrCatHicks/learning-opportunities.git), install the plugin (/plugin install learning-opportunities@learning-opportunities), and restart Claude Code.
• Optional Auto-Prompting: learning-opportunities-auto is available for Linux/macOS, with additional setup for Windows users.
• Prerequisites: A Claude Code environment.

Highlighted Details

• Exercise Types: Includes "Prediction → Observation → Reflection," "Generation → Comparison," "Trace the path," "Debug this," "Teach it back," and "Retrieval check-in."
• Learning Science Foundation: Exercises are grounded in established cognitive and educational psychology principles, designed to enhance retention and understanding.
• Measurement Playbook: A companion guide (MEASURE-THIS.md) offers tools for teams to measure the impact of using this skill, including survey items and guidance on results interpretation.
• Customization: The skill is adaptable, allowing users to incorporate their expertise, adjust trigger conditions, add project-specific examples, and modify exercise caps.

Maintenance & Community

Developed by Dr. Cat Hicks, a psychological scientist specializing in software teams and AI-assisted work. The project emphasizes open science and provides links to research supplements, a newsletter ("Fight for the Human Author"), and consulting services. No specific community channels like Discord or Slack are listed.

Licensing & Compatibility

Licensed under a Creative Commons Attribution 4.0 International License (CC BY-SA 4.0). This license permits commercial use and distribution, provided attribution and share-alike conditions are met.

Limitations & Caveats

The interactive style, where Claude pauses and waits for user input, may feel counterintuitive or frustrating compared to typical fluent AI interactions. Windows users require extra steps for the optional automatic prompting feature. The skill is presented as an experimental add-on for Claude Code.