Skip to main content
AdOpenFree logoPromote your productReach more potential users and drive product growth and revenue.Advertise
Favicon of unslop

unslop

Free Listing

unslop is a Claude Code and Codex skill that audits AI-looking UI, proposes a product-driven direction, and fixes screens pass by pass. It supports frameworks such as Next.js, React, Vue, Svelte, Tailwind, and plain HTML/CSS.

Visit heydamir/unslop

Overview

unslop is a skill for Claude Code and Codex that turns vibecoded, AI-looking UI into a designed interface with its own character. It finds common slop patterns such as purple gradients, glows, emoji labels, shouty copy, and uneven spacing, then sets a direction from the product and fixes the screen pass by pass like a designer doing a redline review. It works across Next.js, React, Vue, Svelte, Tailwind, and plain HTML and CSS, on dashboards, apps, landing pages, and live sites when only a link is available.

Key Features

  • Audit lists every sign of slop with where it lives.
  • Direction derives character, world, color moments, and a signature detail from the product.
  • Passes adjust spacing, color, icons, copy, type, layout, details, visuals, and motion in order.
  • Works with Next.js, React, Vue, Svelte, Tailwind, or plain HTML and CSS, and handles dashboards, apps, landing pages, and live sites via a link.
  • Can generate visuals through image services when a hero needs a scene or object, showing the prompt and asking before every paid generation.
  • Story video creates a square video under 10 seconds of the screen unslopping itself, requiring Playwright and ffmpeg.

Use Cases

  • Redesign a vibecoded dashboard so it reads as a designed product rather than an AI template.
  • Give the hero of a live website a product-specific visual direction and targeted fixes.
  • Improve a landing page and produce a short story video of the transformation for posting.
  • Audit an existing screen and receive small refinements if it is already good rather than a full redesign.

Getting Started

  • For Claude Code, clone the repository and copy skills/unslop into ~/.claude/skills/ or into .claude/skills/ for a single project.
  • For Codex, install the skill from the repository or copy it into ~/.agents/skills/ or .agents/skills/ inside a single project.
  • Invoke it in Claude Code with /unslop or in Codex with $unslop, or ask in your own words so the agent picks it up.
  • To snapshot a live site or create a story video, install Playwright and Chromium from the project directory: npm install -D playwright and npx playwright install chromium.

Deployment & Requirements

  • Playwright and its Chromium browser are required to snapshot a live site or create a story video.
  • Story videos also require ffmpeg installed and available on PATH.

Before You Adopt

  • License: CC-BY-4.0. Review its terms before using, modifying, or distributing the project.
  • Story video generation depends on ffmpeg being installed and available on PATH.
  • Live-site snapshots and story videos require Playwright and its Chromium browser.
  • Paid image generation is guarded by showing the prompt and asking before each paid generation.

Comments

Sign in to leave a comment.

More like unslop

Favicon of geo-sleuth

geo-sleuth

Free ListingStars: 4.8K

Locate a photo’s camera position and show the evidence

Agent Skills & Tool Libraries

An agent skill that finds where a photo was taken using OpenStreetMap geometry, elevation skylines, satellite imagery and street view, then returns a camera position, heading and evidence image.

Favicon of text-to-cad

text-to-cad

Free ListingStars: 23.4K

Give your agent CAD superpowers.

Agent Skills & Tool Libraries

text-to-cad is an open-source plugin that gives AI agents local workflows for generating 3D models as STEP, GLB, STL, or 3MF files, running design-for-manufacturing checks, producing engineering drawings, and connecting to fabrication

Favicon of 3D Model Anything

3D Model Anything

Free ListingStars: 1.3K

Capture rooms with iPhone LiDAR as editable Three.js scenes

Agent Skills & Tool Libraries

3D Model Anything captures iPhone LiDAR and photos, reconstructs editable Three.js rooms with an AI coding agent, and lets you explore the results in a browser.