What problem does it solve? LLM-generated code often suffers from overcomplication, unrequested features, silent assumptions, and sweeping edits that touch unrelated code. This Skill provides behavioral guidelines that keep AI-assisted coding focused, minimal, and verifiable across the VSmartwatch project. ## Core Features & Use Cases - Think Before Coding: Forces explicit statement of assumptions, presents multiple interpretations instead of silently picking one, and stops to ask when requirements are unclear. - Simplicity First & Surgical Changes: Blocks speculative abstractions and unrequested features, and restricts edits to only what the task requires while matching existing codebase style. - Goal-Driven Execution: Transforms vague tasks into verifiable success criteria (e.g., "write tests for invalid inputs, then make them pass") with step-by-step verification plans. - Use Case: When asked to add validation to a login endpoint in a FastAPI or Express backend, the AI defines testable criteria (reject empty email, invalid format, short password), avoids adding unrequested rate limiting, and matches existing code conventions. ## Quick Start Apply the karpathy guidelines while refactoring this Flutter screen and state your assumptions before making any changes.
