From impeccable-rails
Check a UI design against the Design-for-Developers DO/DON'T rulebook — use when asked to review, critique, lint, or sanity-check a design, component, CSS/tokens, or screenshot for concrete rule violations (contrast below 4.5:1, line-height too loose, button text under 14px, arbitrary spacing off the baseline, ALL-CAPS body, underlined links, same font for headers and body, oversized/wrapping headers, misused badges/toasts, weak hierarchy). Complements impeccable and design:design-critique by contributing this specific numeric rubric, not replacing their judgment.
How this skill is triggered — by the user, by Claude, or both
Slash command
/impeccable-rails:lintThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run the book's concrete DO/DON'T rules against a design and report violations. This is a **rubric**,
Run the book's concrete DO/DON'T rules against a design and report violations. This is a rubric,
not a critique engine: it catches measurable rule breaks. For taste, layout, and holistic critique,
defer to impeccable / design:design-critique — run this alongside them, not instead.
references/do-dont.md — the full rule table (id · category · severity · check · fix).impeccable-rails:reference skill
(tokens.json) so lint and scaffold agree.A findings list, grouped by severity:
For each finding: id · the offending value · the rule · the fix (with the correct value from the
reference). End with a one-line tally (N errors, M warnings, K info). If nothing is wrong, say so —
don't invent findings.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
npx claudepluginhub bluefenixproductions/impeccable-rails --plugin impeccable-rails