More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    UISlop: Frontend Vibecode & AI-UI Linter

    1

    Static AST/regex linter for React, Next.js, Vue, Svelte, and HTML that catches 16 vibecoded AI frontend tells and repairs them.

    $5.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    UISlop: Frontend Vibecode & AI-UI Linter

    UISlop: Frontend Vibecode & AI-UI Linter

    Example session with this skill installed

    Audit our Next.js & Tailwind frontend in src/ for vibecoded AI tells (indigo gradient defaults, fake Unsplash avatars, dead href="#" links, and Lucide icon boxes) and refactor it to Grade A.

    • Read your context and instructions
    • Compiled the uislop
    • Generated the UI component

    UISLOP v0.1.0 · Static Vibecode & AI-UI Slop Auditor
    Target : src/ (4-repo 687-file benchmark)
    Files scanned : 1 UI files · 24 findings across 6/6 categories
    Human-Craft Score : 0/100 (Grade F) -> 98/100 (Grade A) after surgical refactor
    Palette Entropy : 20/21 chromatic tokens in indigo/violet/purple (95.2%) -> 0.0%

    [UIS001] LandingHero.tsx:13 · Multi-stop indigo/violet/purple/pink AI gradient
    [UIS006] LandingHero.tsx:11 · Hero eyebrow rounded-full pill badge with <Sparkles>
    [UIS013] Testimonials.tsx:16 · Hardcoded stock avatar URL (images.unsplash.com/photo-1534528741775)
    [UIS015] Footer.tsx:48 · Dead anchor link href="#"

    LandingHero.tsx

    TSX · React component

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Ask Claude Code, Cursor, or v0 for a dark SaaS landing page and you almost always get the same 6 fingerprint clusters: from-indigo-500 via-purple-500 to-pink-500 bg-clip-text text-transparent hero text, a rounded-full <Sparkles /> eyebrow pill badge above the <h1>, 3 identical glassmorphism cards with Lucide icons inside bg-indigo-500/10 boxes, blur-3xl rounded-full floating background orbs, hardcoded images.unsplash.com/photo-1534528741775-53994a69daeb stock avatars next to "Trusted by 10,000+ developers", and dead href="#" / onClick={() => {}} buttons.

    UISlop pairs a zero-dependency Node.js static analyzer (npx uislop / scripts/uislop.mjs) with a surgical remediation workflow:

    What it does

    • 16 Deterministic Static Rules (UIS001-UIS016) Across 6 Categories:
      • Palette & Gradient Defaulting (cap 24 pts): Multi-stop indigo/purple/pink gradients (UIS001), gradient text clipping (UIS002), indigo/violet/purple dominance over 60% of chromatic tokens (UIS003), and blur-3xl rounded-full glow orbs (UIS004).
      • Icon & Pill Badge Cliches (cap 18 pts): Lucide <Sparkles>/<Zap>/<Rocket>/<Shield>/<Brain>/<Cpu> (UIS005), eyebrow pill badges (UIS006), and 3+ icon-in-tinted-box cards (UIS007).
      • Bento & Glass Card Monotony (cap 14 pts): Repetitive backdrop-blur bg-white/5 border-white/10 rounded-2xl wrappers (UIS008) and hover:-translate-y-1 hover:scale-105 card physics (UIS009).
      • Formulaic AI Hero Copy (cap 14 pts): Overused AI hero superlatives (UIS010), unrequested em-dashes in JSX text (UIS011), and "Ready to transform your workflow?" CTAs (UIS012).
      • Fake Social Proof & Avatars (cap 22 pts): Hardcoded images.unsplash.com, pravatar.cc, randomuser.me, and placehold.co URLs (UIS013) and unverified "10,000+ happy users" / "99.9% uptime" claims (UIS014).
      • Hollow Wiring & Dead Links (cap 20 pts): Dead href="#" anchors (UIS015) and no-op onClick={() => {}} / console.log handlers (UIS016), with automatic <pre>, <code>, <svg>, and comment masking.
    • Corpus-Normalized Human-Craft Score (0-100): Computes palette hue entropy across all Tailwind chromatic utilities and normalizes rule penalties by sqrt(4 / filesScanned) so large multi-hundred-file apps are scored fairly.
    • Verified on 687 Real UI Files Across 4 Codebases: rajesh-portfolio (100/100 Grade A), OpenCode-App 536 files (97/100 Grade A), Princeportfolio 129 files (65/100 Grade C), and vibecoded-landing.tsx (0/100 Grade F).

    Requires Node.js to run scripts. Cannot audit compiled CSS, obfuscated builds, or images. Analysis is limited to the 16 defined UIS rules.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 11 days ago

    Requires

    node

    Permissions

    Terminal / Shell
    Listed11 days ago

    What's inside

    Frequently Asked Questions

    uislop scans your source code for specific design patterns, placeholder content like 'Acme Corp,' and formulaic AI writing that signals a project was generated rather than built, giving you a 'Human-Craft Score' to track improvement.

    The skill includes a Node.js-based static analysis script, a comprehensive library of UI cliché definitions, and a guide on refactoring your code to achieve a high Human-Craft Score.

    Yes, uislop is designed to run locally on your source files, making it the perfect final step after using generative tools like v0, Bolt.new, or Claude Engineer to ensure the output looks professional and bespoke.

    uislop is a deterministic linter that uses static analysis to find exact file and line citations; unlike an LLM, it won't hallucinate or overlook the very tropes it was trained on.

    It requires a Node.js environment and works with React, Next.js, Vue, Svelte, Astro, and Tailwind CSS projects; however, it cannot audit compiled production builds or obfuscated code.

    New in Frontend & Web Apps