What problem does it solve?
AI-generated frontend code often includes generic, inconsistent design patterns that betray its origin, create visual slop, and deviate from your codebase's established design standards, making code harder to maintain and less professional for end users.
Core Features & Use Cases
- 12 Deterministic Detection Rules: Catches specific AI tells including shadow variant explosion, generic violet-to-pink gradients, lorem ipsum placeholders, inline styles, off-grid spacing, off-palette colors, Tailwind class bloat, missing alt text, leftover console.log statements, emoji in headings, and default transition-all usage.
- Codebase-Aware Scanning: When paired with a dragoon.json manifest from a prior scan, rules adjust to match your project's unique spacing grid and color palette for more accurate, relevant findings.
- Actionable Findings: Every detected issue includes the exact file, line, column, offending code snippet, explanation of why it is a problem, and a concrete fix, with severity ratings (low/medium/high) for prioritization. Ideal for PR reviews, pre-ship quality checks, and design hygiene audits.
Quick Start
Use the slop skill to scan your project's UI code for AI-generated design patterns and get specific, actionable fixes for every issue found.