Skip to content
SkillScript-led4,076 tokens in use

Oil UI

A UI/UX design workflow for coding agents: one clear visual direction, optional three-way comparison pages, and polish checked against real screenshots.

3,593 stars on the repository MIT Licenseoil-oil/oil-ui
npx skills add oil-oil/oil-ui
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Oil UI costs

272tokens idle
4,076tokens in use
56,820tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Script-led

Script-ledMixedModel-led

Most of this skill is scripts, so it behaves much the same whichever model runs it.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: assets/vendor/alpine.min.js, scripts/build_explorer.py, scripts/check_update.py, scripts/check_update.sh, scripts/chrome.mjs, scripts/curate-media.mjs, scripts/measure_reference.py, scripts/oiloil-image.mjs, scripts/prep-image.mjs, scripts/shoot.mjs.

Touches

shelldeclared and detectednetworkdetectedwrites filesdetectedcredentialsdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter8/15
  • Structure8/25
  • Examples6/15
  • Output0/15
  • Size13/15
  • Resources6/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Fair
  • Runs shell commands−2

    does this on the machine, as the frontmatter declares

    SKILL.md, line 27
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 27
  • Minified code−10

    a line of 6409 characters: minified or generated, so it cannot be read

    assets/vendor/alpine.min.js, line 2
  • Reaches the network−10

    does this on the machine, and the frontmatter does not say so

    scripts/build_explorer.py, line 19
  • Reads credentials−15

    does this on the machine, and the frontmatter does not say so

    scripts/check_update.py, line 99

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity100
  • Maintenance100
  • Community44
  • Maturity12
  • Momentum40

Oil UI guides an agent through designing, refining and reviewing websites, apps, dashboards and components. It sets a one-sentence visual north star first, then shapes layout, type, color, media and interaction around it, and judges the result from real browser screenshots.

It also builds a side-by-side comparison page of three distinct directions when you ask, flags conflicts between design documents such as DESIGN.md, and runs a read-only review mode that gives evidence and advice without touching files. The skill text is written in Chinese and replies in the language you use.

What You Get

  • A single visual direction stated up front, or three clearly different directions on one comparison page when you ask
  • Desktop and mobile screenshots of the built page, inspected before the result is handed over
  • A read-only review with evidence and ranked advice when you only want critique
  • A delivery note listing what was built, what was not done and up to three suggested next steps

Common Use Cases

  • Design a new landing page or product screen from a blank start
  • Compare three styles for the same page and pick one
  • Polish an existing dashboard or component without changing its logic
  • Recreate a page from a screenshot with a mobile layout
  • Review a homepage and get written changes without edits

What Makes It Unique

The skill separates how an interface looks from whether its tasks work, and gives each its own acceptance step. Page type decides what the first screen must contain: persuasion pages follow a chosen opening layout, while tool and browse pages put the first row of content on screen one. Explicit comparisons require the three directions to share at most one trait between any two, and a separate reviewer without build history scores the result.

Under The Hood

SKILL.md is a router that reads a short light-path file for common tasks and loads other references only when a situation calls for them. Its frontmatter allows shell calls to a version-check script, which reads a public version list on the Oil UI site, caches the result under a state folder in your home directory and only notifies you; it updates nothing unless you ask. Optional helper scripts in the folder take screenshots with a local Chrome, curate and prepare media, build the comparison page, and call an image-generation service that spends credits when you have authorized it. The agent writes comparison pages, screenshots and intermediate files to a working folder named .oil-ui in your project, and final code to your project files.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers