skills-impeccable/i-polish/SKILL.md
Use when the user says: "polish this UI", "final pass", "pixel-perfect", "normalize the design", "make it consistent", "align with design system". Final quality pass for alignment, spacing, consistency, and design system alignment.
npx skillsauth add NodeJSmith/Claudefiles i-polishInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Read ${CLAUDE_HOME:-~/.claude}/skills/i-frontend-design/SKILL.md for design principles, anti-patterns, and the Context Gathering Protocol. Follow the protocol before proceeding — if no design context exists yet, you MUST run /i-teach-impeccable first. Additionally gather: quality bar (MVP vs flagship).
Perform a meticulous final pass to catch all the small details that separate good work from great work. The difference between shipped and polished.
Before polishing, understand the system you are polishing toward:
If a design system exists, polish should align the feature with it. If none exists, polish against the conventions visible in the codebase.
Understand the current state and goals:
Review completeness:
Identify polish areas:
CRITICAL: Polish is the last step, not the first. Don't polish work that's not functionally complete.
After analyzing the current state, present your proposed changes to the user:
Then STOP and confirm before implementing:
AskUserQuestion:
question: "Here's what I'll check and fix in this polish pass. Proceed?"
header: "Confirm"
options:
- label: "Implement"
description: "Looks good — go ahead and make these changes."
- label: "Refine scope"
description: "I want to adjust what's included before you start."
- label: "Challenge this first"
description: "I'll run /mine.challenge against your proposal before we proceed."
- label: "Stop here"
description: "Don't implement anything. The proposal is in this conversation only."
If "Implement" → proceed to implementation below. If "Refine scope" → ask what to change, update proposal, re-confirm.
<!-- CHALLENGE-CALLER -->If "Challenge this first" → invoke /mine.challenge inline against the proposal, read findings, revise proposal, re-present this gate.
If "Stop here" → end the skill.
Work through these dimensions methodically. The standard checks (grid alignment, WCAG contrast, 44px touch targets, removing console.logs/dead code, alt text, no layout shift) you already know — apply them. The items below are the ones worth spelling out:
prefers-reduced-motion.IMPORTANT: Polish is about details. Zoom in. Squint at it. Use it yourself. The little things add up.
NEVER:
Before marking as done:
Remember: You have impeccable attention to detail and exquisite taste. Polish until it feels effortless, looks intentional, and works flawlessly. Sweat the details - they matter.
After polishing, ensure code quality:
After implementation, summarize in conversation:
development
Use when the user says: "humanize this", "unslop this", "de-slop this", "fix AI writing", "remove AI tells", "clean up AI prose". Edits prose to remove AI writing patterns and add human voice. Analyzes first, then asks how to fix. Prose complement to mine.clean-code.
development
Use when the user says: "why is this code like this", "why does this exist", "why was this built this way", "decision rationale", "what's the history behind". Decision archaeology — reconstructs historical rationale from evidence, not speculation.
development
Use when the user says: "how does X work", "walk me through", "explain this subsystem", "explain how", "trace the flow". Complexity-adaptive subsystem explanation — builds mental models conversationally, not documentation artifacts.
development
Use when the user says: 'create an issue', 'file an issue', 'open an issue', 'write an issue', 'new issue for this'. Codebase-aware issue creation — investigates the code to produce well-structured issues with acceptance criteria, affected areas, and enough detail for automated triage.