What problem does it solve?
This Skill solves the common pain of working with codebases full of shallow, tightly-coupled modules that are hard to test, hard to maintain, and difficult for AI tools to navigate. It surfaces hidden architectural friction and proposes targeted refactors to turn shallow modules into deep, high-leverage ones, improving both developer productivity and AI navigability.
Core Features & Use Cases
- Architectural Friction Detection: Scans codebases to identify shallow modules (where interface complexity nearly matches implementation complexity), tightly-coupled modules that leak across seams, and hard-to-test areas.
- Domain-Aligned Recommendations: All suggestions use the project's existing domain language from CONTEXT.md and respect recorded architectural decisions (ADRs) to avoid conflicting with prior choices.
- Use Cases: Ideal for refactoring legacy codebases, improving test coverage, consolidating redundant modules, and making codebases more navigable for both human developers and AI coding assistants.
Quick Start
Use the improve-codebase-architecture skill to scan the current project codebase and list the top 3 highest-impact deepening opportunities, aligned with the domain terms defined in the project's CONTEXT.md.