Skip to content
SkillScript-led2,784 tokens in use

impeccable

A design toolkit for coding agents: 24 commands to build, critique, audit, polish and animate a frontend, driven by PRODUCT.md and DESIGN.md context files.

113,983 stars on the repository Apache License 2.0pbakaus/impeccable · plugin/skills/impeccable
npx impeccable install
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing impeccable costs

49tokens idle
2,784tokens in use
90,206tokens, 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: scripts/impeccable, scripts/live-browser-dom.js, scripts/live-browser-ignores.js, scripts/live-browser-session.js, scripts/live-browser.js, scripts/modern-screenshot.umd.js.

Touches

shelldetectednetworkdetectedwrites filesdetected

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 →
Good
  • Frontmatter11/15
  • Structure25/25
  • Examples0/15
  • Output15/15
  • Size15/15
  • Resources9/15

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−5

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

    scripts/impeccable, line 19
  • Writes files−8

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

    scripts/impeccable, line 98
  • Reaches the network−10

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

    scripts/impeccable, line 193
  • File that was not read−15

    a file the skill ships that was not read, so what it does is unknown

    scripts/impeccable.cmd
  • Minified code−10

    a line of 9212 characters: minified or generated, so it cannot be read

    scripts/modern-screenshot.umd.js, line 1

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity100
  • Maintenance100
  • Community68
  • Maturity36
  • Momentum40

Impeccable gives a coding agent a design vocabulary. You name a command such as critique, audit, polish, bolder or animate and a target, and the agent loads a playbook written for that one move. Context files PRODUCT.md and DESIGN.md hold who the product is for and how it looks, so every command stays on-brand.

It covers websites, landing pages, dashboards, components, native iOS and Android screens and design systems. A launcher runs a self-contained engine binary that loads project context, takes screenshots and checks the result.

What You Get

  • Commands for building, critiquing, auditing, polishing, animating, hardening and adapting a frontend
  • A PRODUCT.md and DESIGN.md that record the product and its visual system
  • Playbooks for each command, plus guidance for native iOS and Android
  • Screenshots of the rendered result at shipping sizes, taken by the bundled launcher

Common Use Cases

  • Reviewing an existing page with critique and audit
  • Building a new landing page with a committed visual direction
  • Running a final polish and harden pass before launch
  • Capturing an existing visual system in DESIGN.md

What Makes It Unique

Most design skills give general taste advice. Impeccable splits design work into named commands, each with its own playbook, and ties them to context files the agent reads every run. It treats the brief as the authority, separates refinement from redesign, and sets ambition by surface mode: bold for marketing pages, precise for app UI.

Under The Hood

The SKILL.md routes a request to a reference file. A shell launcher runs a self-contained engine binary that sits beside it, or downloads one into the user’s home directory on first run and checks it against a listed SHA-256 checksum before running it. The engine loads context, takes screenshots and runs the design detector. Live mode opens a browser overlay served from a local dev server. The agent reads and edits project source files and writes context files such as PRODUCT.md, DESIGN.md and an .impeccable folder.

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