What problem does it solve? Codebases naturally grow over time, accumulating complexity, maintenance burden, and bugs. This Skill counters that trend by evaluating every change based on whether it reduces the total amount of code in the final result, rather than minimizing immediate effort. ## Core Features & Use Cases - Deletion-Biased Evaluation: Applies three core questions to every change—what is the smallest codebase that solves this, does the result reduce total code, and what can be deleted. - Reference Mindsets: Loads philosophical frameworks from the references directory (such as simplicity-vs-easy, data-over-abstractions, and PAGNI) to ground simplification decisions before starting work. - Anti-Pattern Detection: Rejects common excuses for adding code, including status quo bias, speculative flexibility, and unnecessary abstraction layers. - Use Case: When asked to add validation to five forms, instead of creating six new files, the Skill guides you to consolidate forms and logic, potentially reducing the codebase by 150 lines while preserving functionality. ## Quick Start Ask the AI to reduce entropy in a specific module by applying the reducing-entropy skill and loading a reference mindset before proposing deletions.
