What problem does it solve?
This Skill solves the common pain of codebases filled with shallow, tightly coupled modules that are hard to test, hard for AI agents to navigate, and require changes across dozens of files for even small feature updates. It eliminates guesswork in architecture refactoring by providing a structured, vocabulary-consistent process to identify high-impact consolidation opportunities.
Core Features & Use Cases
- Shallow Module Detection: Uses the deletion test and organic codebase exploration to identify modules where the interface is nearly as complex as the implementation, indicating they are not earning their keep.
- Deepening Opportunity Scoring: Ranks refactoring candidates by their potential to improve testability, reduce change locality, and increase interface leverage for both human developers and AI agents.
- ADR-Aligned Recommendations: Ensures proposed refactors do not conflict with existing architecture decision records, only surfacing changes that are worth revisiting past prior decisions.
- Use Case: For a team struggling with a codebase where updating a single user-facing feature requires modifying 12 separate utility modules, this Skill surfaces the opportunity to merge those utilities into a single deep module with a clear, small interface, cutting change spread and test maintenance in half.
Quick Start
Use the improve-codebase-architecture skill to analyze the current project codebase and list the top 3 deepening opportunities that would reduce the number of files modified for a typical feature update.