What problem does it solve? Pure AI delegation speeds up execution but erodes the user's own skill formation, leaving them unable to understand, debug, or reproduce AI-generated code. This Skill steers the assistant toward interaction modes that keep the user learning instead of passively receiving output. ## Core Features & Use Cases - Interaction Mode Selection: Distinguishes low skill-formation patterns (pure delegation, AI-driven debugging) from high skill-formation modes (conceptual inquiry, generation-then-comprehension, hybrid code-explanation). - Learning Tier Decisions: Matches the assistance level to the user's familiarity, from manual-first attempts for new concepts to full delegation for mastered domains. - Decision Matrix & Red Flags: Provides concrete rules for scenarios like debugging, architecture choices, and repeated errors, plus warning signs of over-reliance. - Use Case: A user learning a new framework asks for a fix. Instead of patching the code, the assistant asks what they have tried, explains the underlying mental model, and prompts them to explain the solution back. ## Quick Start Ask the assistant to help you learn a new concept using the interactive-learning approach rather than just fixing the code for you.
