From sdlc
UI quality audit — inspects computed CSS across typography, spacing, color, layout, interaction, accessibility, and polish. Produces letter grades per category. Requires Preview MCP tools. Trigger: 'audit the design', 'check UI quality', 'grade the frontend', 'design review'.
How this skill is triggered — by the user, by Claude, or both
Slash command
/sdlc:design-auditThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Log skill invocation:
Log skill invocation:
AUDIT_SCRIPT=$(find . -name "audit-trail.sh" -path "*/sdlc/*" 2>/dev/null | head -1)
[ -z "$AUDIT_SCRIPT" ] && AUDIT_SCRIPT=$(find "$HOME/.claude" -name "audit-trail.sh" -path "*/sdlc/*" 2>/dev/null | sort -V | tail -1)
bash "$AUDIT_SCRIPT" log review sdlc:design-audit started --context "$ARGUMENTS"bash "$AUDIT_SCRIPT" log review sdlc:design-audit completed --context="<summary>"Perform a structured 80-item design audit powered by browser CSS inspection. Each of 7 categories gets a letter grade (A-F).
/sdlc:design-audit to audit the current UIpreview_inspect for computed values, not guessing.quality/proof/design-audit.json| Category | Items | Key tools |
|---|---|---|
| Typography | 12 | preview_inspect for font-size, line-height, font-weight |
| Spacing | 10 | preview_inspect for padding, margin, gap |
| Color | 12 | preview_inspect for color, background-color + contrast calc |
| Layout | 12 | preview_resize + preview_screenshot for responsive |
| Interaction | 12 | preview_eval for hover/focus states |
| Accessibility | 12 | preview_snapshot for ARIA, roles, labels |
| Polish | 10 | Mix of all tools |
| Score | Grade |
|---|---|
| >= 90% | A |
| >= 80% | B |
| >= 70% | C |
| >= 60% | D |
| < 60% | F |
Configurable minimum grade in sdlc.config.json (default: no category below C).
.quality/proof/design-audit.json — per-item scores, per-category grades, overall grade.quality/proof/screenshots/*.png — viewport screenshots and issue documentationThe ship skill includes the grade table and screenshots in the PR description under a ## Design Audit section.
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 jerrod/agent-plugins --plugin sdlc