{
  "id": 18276624,
  "title": "A Wardrobe Raid for Sparkle Emoji, an Oompa Loompa Shift for Every Ticket, and a Draw 25 Card Called Accept All",
  "url": "https://urgent.news/2026/11/05/a-wardrobe-raid-for-sparkle-emoji-an-oompa-loompa-shift-for-every",
  "topic": "culture",
  "section": "Culture",
  "published": "2026-11-05T08:36:43.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/thereactnativerewind/a-wardrobe-raid-for-sparkle-emoji-an-oompa-loompa-shift-for-every-ticket-and-a-draw-25-card-1pdj"
  },
  "original_language": "en",
  "account": "In the world of coding, a new wardrobe raid has emerged to ensure sparkling code quality. Trinny and Susannah may judge fashion, but now AI-built screens require a wardrobe raid of their own. Sinan Yorulmaz has created an open-source package called ai-design-check to enforce 95 design rules and test 57 of them automatically. React, Vue, Svelte, HTML, SwiftUI, and Jetpack Compose apps all pass through this wardrobe raid, with findings labeled by file, line, and rule. The final decision rests with a human or agent to make the necessary changes, preserving features, data, and copy.\n\nEnter the factory of coding, powered by SuperPlane 1.0. Darko Fabijan has built this software factory that transforms tickets into tested code changes. The factory scores tasks for clarity, complexity, and verifiability before any agent begins work. Approved tasks run in parallel environments, with the chosen model, and every change is accompanied by evidence, such as passing checks and screenshots. The whole team shares a board displaying the status of tasks, failures, and costs.\n\nSuperPlane is released under the Apache 2.0 license and can be installed on a single VM with Docker Compose. With concurrent runners, each requiring 2 vCPUs and 8 GB of memory, the factory is cost-effective to open but not to fully staff. Currently, 52% of pull requests merged last month went through SuperPlane, tackling tasks that might have otherwise gone unaddressed.\n\nGitHub Agentic Workflows offer a new way to write CI chores in Markdown, turning them into Actions YAML that GitHub Copilot, Claude Code, or Codex triage. The process is read-only and sandboxed by default, preventing agents from merging their own homework. GitHub Verified MCP Connectors allow integration with Slack, Figma, Notion, Sentry, Linear, and 21 more services, streamlining the workflow across various platforms.\n\nCopilot in VS Code, Universe Edition, provides cloud sessions that follow users from phone to desk, an Agent Host that keeps running after closing the window, and experimental computer use. Standups can now start a PR directly, making the development process more efficient.\n\nGitHub Copilot Server-to-Server Auth enables GitHub Apps to call the Copilot SDK on behalf of an organization without requiring a signed-in human, billed to the organization. Replit Cloud Agents allow any team member on a Pro or Enterprise plan to describe a change, and the agent clicks through it in a sandbox, opening a PR with a video attached. Marketing teams can now ship code while engineering still holds the merge button.",
  "summary": "What Not to Ship, Susannah! Trinny and Susannah never asked what you fancied wearing. They opened the wardrobe, held up the worst offender and told you exactly why it had to go. AI-built screens need the same treatment: sparkle emoji, tracked uppercase labels, 10px text and a gradient on everything. Sinan Yorulmaz ( @sinanyorulmazz ) of Enmera has built that wardrobe raid as an open-source…",
  "key_points": [
    "AI-built screens require 95 design rules in wardrobe raid",
    "SuperPlane factory transforms tickets into tested code changes",
    "GitHub Agentic Workflows write CI chores in Markdown"
  ],
  "editors_take": "Automating design enforcement and ticket processing shifts the developer's role from manual execution to high-level oversight, allowing non-engineers to ship code while maintaining strict quality standards.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}