agentleFS
Sign inStart free
SubliminalCodingSubliminalCoding/bilbroswagginz-skills/skills/qwen-ux-maximizer/SKILL.md

qwen-ux-maximizer

Skill0 starsChanged 47 days agoMITView on GitHub

Local-Qwen UI/UX pass (token-free), heavily user-gated because taste is not a small- model job. Mechanical form of /ux-maximizer: fix only CONCRETE, checkable UI issues (contrast, missing labels, broken layout, errorless failures) under the gates; for anything about look/feel/"is this premium", screenshot it and STOP for the user. Use when you want /ux-maximizer's job done on a local model with no API tokens: basic UI hardening.

The file

Create a free account to mount
Files
SKILL.md

qwen-ux-maximizer (local UI/UX; taste stays with the user)

The Qwen-run version of /ux-maximizer, with an honest limit: a small local model cannot judge taste. So it fixes only the OBJECTIVE issues and hands everything subjective to the user.

QWEN LOCAL RULES (full text: ../qwen-forge/references/QWEN-MODE.md)

Requires qwen-forge installed alongside this skill (it holds the shared rules).

  • Builder, not architect. ONE change at a time; build → tests → lint; green → commit; red → revert, one retry, else STOP. Never weaken tests.
  • Unclear / judgment / outward → DECISIONS.md + STOP.
  • Local endpoint required: a local OpenAI-compatible endpoint (vLLM, SGLang, llama.cpp or Ollama) serving a capable Qwen-class coding model. ONE model loaded, work SEQUENTIALLY, leave memory headroom on unified-memory machines.

Steps

1. CHECKABLE FIXES ONLY. Run ux-scan / accessibility checks and fix issues that are
   OBJECTIVE and verifiable, one at a time under the gates:
     - color contrast below standard
     - missing alt text / aria-labels / form labels
     - broken or overflowing layout, off-screen elements
     - actions that fail with no message to the user
     - obvious mobile breakage (unreachable buttons, tables overflowing)

2. TASTE → USER. Anything about look, feel, "make it premium", style, palette, vibe,
   layout aesthetics → take a screenshot, write what you'd change in DECISIONS.md, and
   STOP. Do NOT invent a design direction. You cannot judge whether it looks good.

3. NEVER run idea-wizard locally to brainstorm a redesign: overbuild risk, and it's
   the user's creative call.

Rule

  • The hard line: objective accessibility/correctness UI bugs = you fix them. Subjective beauty/taste = screenshot + STOP. When unsure which side a thing is on, treat it as taste.

Distilled from /ux-maximizer. Its PROMPTS.md holds the full wow-factor idea-wizard pipeline for when the user or Claude is driving the design.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free