designlint-skill

Name: designlint-skill
Author: rgranet

rgranet March 12, 2026

1 copies 0 downloads

Stop AI-generated interfaces from all looking the same. DesignLint forces AI agents to understand the user, make 9 deliberate design decisions, and critique the result before writing code. Two modes: Create from scratch, or Audit and fix existing UIs.

DesignLint

Force distinctive, human-feeling UI/UX design before any AI agent writes code. Audit and improve existing interfaces.

By Ruben Granet · v1.2.0 · MIT License

The Problem

AI coding agents converge. Left unconstrained, they always produce the same thing:

UI: Inter font · Purple/blue gradients · White background · rounded-xl cards · Opacity fade-ins

UX: Tab bar (5 icons) · List → Push detail · Modal for everything · Green toast on success · 3-screen onboarding carousel

Voice: "Submit" · "An error occurred" · "No items found" · "Success!" · (spinner)

This isn't a style preference — it's a structural failure. Every AI-generated interface looks, feels, and sounds like every other AI-generated interface.

DesignLint breaks that convergence with two modes:

Create — Understand the user, make 9 deliberate design decisions, resolve a creative tension, and critique the result before shipping.
Audit — Diagnose an existing interface against the same standards, score it, and produce a concrete improvement plan.

Two Modes

Create Mode

"Build me a task management app"

The agent stops before writing code. It identifies the user and their emotional state, picks a creative tension, selects an archetype as a departure point (not a template), makes 9 design decisions covering aesthetics through voice, and runs a human critique before shipping.

Audit Mode

"What's wrong with this UI?" / "Improve this interface" / "Review this design"

The agent documents what exists, runs the full convergence scan (UI, UX, and human), scores the design 1–5, produces an improvement brief with the top 3 highest-impact changes, and optionally executes the improvements.

Before / After

Without this skill, an AI agent asked to "build a task management app" produces:

UI:  Inter · #7C3AED primary · white bg · rounded-xl cards · fade-in
UX:  Tab bar (Home, Tasks, Calendar, Search, Profil

Comments

More Agents

View all

agent-memory

Emulo

Mine your Claude Code and Codex logs into a local you.md agent profile.

ohad6k

193

lm-studio

Nyx Local Ai

Local-first AI coding agent for VS Code & Cursor. Ollama, LM Studio & your inference fleet. Cursor-grade agent UX — offline, private, zero token cost.

sthamann

248

Self Learning Skills

A self-improving skill for AI coding agents (Claude Code, Cursor, AGENTS.md): recognize a hard-won golden path in a session and harvest it into a reusable skill/rule for next time.

Kulaxyz

895

agentic-ai

FDEOps

Second brain for Forward Deployed Engineers. Engagement memory + 35 skills across 6 domains, all behind one @fde... Works with any AI coding agent.

suboss87

303

agent-skills

Awesome Gamedev Agent Skills

Game-development Agent Skills for AI coding agents: install once and a master router loads the right skill for your engine and task. 66 original, version-pinned skills (plus a master router) in the portable SKILL.md format that runs across Claude Code, Cursor, Codex, Copilot, Gemini CLI and more, for Godot, Unity, Unreal, web and beyond.

gamedev-skills

301

agents

Honey For Devs

Honey (I Shrunk the AI) by GreenPT: a cross-tool coding skill that cuts AI coding-agent token usage and LLM API costs — write less code, less prose, and denser agent-to-agent handoffs (−53%, lossless in benchmarks) with no loss of quality. Works with Claude Code, Cursor, GitHub Copilot, Codex, Gemini CLI, Windsurf, Cline & Kiro.

Green-PT

177