- Home
- Skills
- Frontend & Web Apps
- frontend dev human engineered
Works with the AI Agents you already use
frontend dev human engineered
Build custom, production-grade frontend components that avoid generic AI design patterns and uniform code defaults.
Free
frontend dev human engineered
Example session with this skill installed
Restyle this generic login form to feel like a high-security terminal for a cybersecurity firm. Avoid rounded corners and trendy gradients. Use a real spacing scale.
- Read your context and instructions
- Compiled the frontend dev human
I have restyled the login interface to move away from generic SaaS defaults. The design now uses a high-density, monospaced aesthetic to reflect a secure terminal environment. I replaced all rounded-xl utilities with hard edges, implemented a strict 4px grid for spacing, and swapped the system font for a professional mono-pair.
Security Terminal UI
| Aspect | Implementation Detail |
|---|---|
| Typography | JetBrains Mono (Headings) / IBM Plex Mono (Body) |
| Palette | Deep Slate (#0F172A), Terminal Green (#22C55E), Alert Amber (#F59E0B) |
| Signature Move | Command-line style focus states with blinking block carets |
| Spacing | Strict 4px/8px increments using Tailwind scale (p-2, p-4, m-1) |
import React, { useState } from 'react';
// Domain-driven naming: AuthTerminal instead of LoginCard
const AuthTerminal = () => {
const [status, setStatus] = useState<'idle' | 'authenticating' | 'error'>('idle');
return (
<div className="bg-slate-950 border border-slate-800 p-6 max-w-md w-full font-mono">
<div className="flex items-center gap-2 mb-8 border-b border-slate-800 pb-2">
<span className="h-2 w-2 bg-green-500 animate-pulse" />
<h1 className="text-slate-200 text-sm uppercase tracking-widest">System.Access_Request</h1>
</div>
<form className="space-y-6">
<div>
<label className="block text-xs text-slate-500 uppercase mb-2">Operator_ID</label>
<input
type="text"
className="w-full bg-slate-900 border border-slate-700 text-green-500 p-2 focus:border-green-500 focus:outline-none"
placeholder="root@internal"
/>
</div>
<div>
<label className="block text-xs text-slate-500 uppercase mb-2">Access_Key</label>
<input
type="password"
className="w-full bg-slate-900 border border-slate-700 text-green-500 p-2 focus:border-green-500 focus:outline-none"
/>
</div>
<button
className="w-full bg-green-600 hover:bg-green-500 text-slate-950 py-2 text-sm uppercase font-bold transition-colors"
>
Execute_Auth
</button>
</form>
{/* Explicit error state handling */}
{status === 'error' && (
<div className="mt-4 border border-amber-900/50 bg-amber-950/20 p-2 text-xs text-amber-500">
[!] AUTHENTICATION_FAILURE: INVALID_CREDENTIALS
</div>
)}
</div>
);
};
export default AuthTerminal;
Next steps
- Review the
AuthTerminalnaming convention to ensure it fits your project's directory structure. - Adjust the
Terminal Greenhex code if it conflicts with your existing brand primary. - Add an
AuthLockoutstate for multiple failed attempts to complete the terminal logic.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop shipping generic, cookie-cutter interfaces that scream "AI-generated." Most LLMs default to the same four visual moves: purple-to-blue gradients, glassmorphism, and identical bento grids. This skill forces the agent to work like a senior frontend engineer who makes deliberate structural and aesthetic choices based on your specific product domain rather than trendy defaults.
What it does
- Eliminates AI tells by avoiding uniform defaults like
rounded-2xlcards and unnecessary scroll-triggered animations. - Enforces domain-driven naming by using variables and components that describe business logic (e.g.,
OrderSummary) instead of JSX shapes (e.g.,CardWrapper). - Defines custom design systems including specific hex palettes, typography pairings, and spacing scales tailored to the content density.
- Prioritizes state handling by building bespoke empty, loading, and error states for async data instead of generic spinners.
- Applies intentional motion where animation clarifies state changes or spatial relationships rather than acting as simple decoration.
How it works
- Context analysis: The agent audits your existing codebase to match spacing scales, naming conventions, and component patterns.
- Intentional planning: It drafts a brief stating the target user, the primary action, and the specific "signature move" for the screen.
- Execution: The agent builds the UI, deliberately avoiding the "AI-built calibration list" of common defaults.
- Peer review: It critiques its own output against senior engineering standards before delivering the final code.
Frameworks & tools
Designed for modern frontend stacks including React, Next.js, and Tailwind CSS. It works within any project using a structured design system or utility-first CSS framework.
Why this beats prompting it yourself
Generic prompts produce generic code. This skill provides a built-in "calibration" layer that identifies and replaces common LLM defaults—like arbitrary Tailwind values and identical padding—with production-grade engineering decisions that survive a real PR review.
Use cases
- De-genericizing landing pages that look like every other SaaS template on the market.
- Building complex dashboards that require high information density without falling back on basic card grids.
- Refactoring legacy components to use domain-specific naming and better state management.
- Styling niche products like security tools or educational apps that need a specific psychological "feel" beyond the standard tech aesthetic.
Known limitations
Does not perform high-level brand strategy or logo design. Relies on the user to provide initial product constraints or domain context for the best results.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Frontend & Web Apps.
Requires initial domain context or existing repo patterns to avoid assumptions. Does not generate brand assets like logos or custom illustrations.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 10 days ago
- Free with an Agensi account