feat: improve skill scores for uncodixify-skill

[rohan-tessl]

Hey 👋 @dazeb

I ran your skills through tessl skill review at work and found some targeted improvements.

Here's the full before/after:

Skill	Before	After	Change
uncodixify	70%	94%	+24%

Changes made

uncodixify — The skill's description was strong on "when to use" but weak on explaining "what it actually does." The judge flagged that concrete actions and outputs were missing, and that trigger terms were too niche to match natural user phrasing.

Changes:

• Rewrote the description field to explicitly state what the skill does: audits component styling against design constraints, flags banned CSS patterns, and enforces a dark restrained aesthetic via an automated validator
• Added natural trigger phrases users would actually type (UI looks generic, fix the styling, Tailwind design system, clean up the CSS, component looks like a template)
• Added a concrete before/after TSX code snippet showing a banned pattern (gradient + glassmorphism) replaced with a compliant alternative
• Added an example of what validator failure output looks like, so the agent knows how to interpret and act on it

---

⚡ Your skills could self-improve on every PR

Right now, skill quality depends on whoever has time to run a manual review. It doesn't have to work that way.

tesslio/skill-review-and-optimize is a free GitHub Action that turns every PR into an automatic skill quality check — with AI-powered rewrites ready to merge in one click.

Here's what happens the moment you add it:

Without the Action	With the Action
Skills drift in quality between manual audits	Every PR gets an instant quality score
Contributors guess whether their edit is an improvement	AI feedback lands in the PR comment before review
Fixing a bad skill means a separate PR, separate effort	Comment /apply-optimize → improvement committed automatically
Quality depends on who has bandwidth	Quality compounds with every merge

• 🚦 Non-blocking by default — purely feedback. No surprise red CI, ever, unless you choose to enforce a score floor
• ⚡ One-click fixes — comment /apply-optimize on any PR and the AI suggestions are committed to the branch automatically

Think of it as a spell-checker for your skills — always on, always improving, costs nothing.

Please check one:

• Yes — open a follow-up PR with the skill review + optimize GitHub Action
• No thanks, I'll manage skill quality manually

---

Honest disclosure — I work at @tesslio where we build tooling around skills like these. Not a pitch - just saw room for improvement and wanted to contribute.

Want to self-improve your skills? Just point your agent (Claude Code, Codex, etc.) at this Tessl guide and ask it to optimize your skill. Ping me - @rohan-tessl - if you hit any snags.

Thanks in advance 🙏