Urgent.News

What's breaking now, across thousands of outlets.

AI

One Vue app, built by AI three times: no skill, vuejs-ai/skills, and vue-enterprise

A few weeks ago I published vue-enterprise , an Agent Skill that teaches Claude Code and GitHub Copilot how I want Vue 3 + TypeScript code written in large, long-lived apps. People asked a fair question: does a skill change anything, now that models write decent Vue on their own? So I measured it. One app, built by an AI agent with no skill, with the community vuejs-ai/skills , and with…

Three Vue apps were built by AI, one with no skill, one using vuejs-ai/skills, and one using vue-enterprise. Each app was tested six times, resulting in twelve apps in total. The apps all performed correctly without requiring any human intervention. The main differences between the apps were in their code structure and functionality.

The no-skill app fetched data on each page and stored it in memory, leading to inconsistencies across different pages and filters. The vuejs-ai/skills app used a cache for server data, ensuring consistency across pages and filters. The vue-enterprise app used Pinia's cache and handled optimistic updates, ensuring that changes were reflected across all pages and filters.

The review scores showed that the vue-enterprise apps performed best in terms of structure, server data handling, and detailed features.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Google Expands Gemini Across Sheets, Search and Developer Tools With Staggered Rollouts

Google is expanding Gemini across consumer, Workspace and developer surfaces, but this is not one uniform launch. The most concrete updates include Gemini in Google Sheets on Android for eligible…

  • Gemini AI integrated into Google Sheets for Android, starting rollout on September 9, 2026
  • Public access to Gemini API through Google AI Studio for developers
  • AI-powered coding assistance in Android Studio for developer workflows

How SteerDistill Shrinks 30‑Billion‑Parameter LLMs to 3 GB Without Losing Accuracy

SteerDistill — How Closed‑Loop Teacher Steering Rescues Quantized LLM Agents Lead “Quantizing a 30‑billion‑parameter agent drops its success rate by 20 %—that’s the headline most engineers see.” Yet…

  • SteerDistill shrinks 30-billion-parameter LLMs to 3 GB footprint
  • Maintains accuracy with closed-loop teacher-steering pipeline
  • Improves inference latency by half and reduces GPU memory use

More from Thursday 12 November →