mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Consolidated UI engineering team plugin: design, review, improvement and performance optimization for any UI surface (web, iOS, Android, desktop), with deep TypeScript/React/Tailwind v4 coverage plus Apple and Android review overlays. Ten specialists (architect, visual reviewer, anti-slop auditor, accessibility, motion, responsive, perf engineer, TypeScript engineer, verifier, team lead), each dispatched on the model the session chooses. Four user-invoked skills: design-ui (new UI from scratch), review-ui (adaptive specialist review with delta reporting against a review ledger), improve-ui (full team pass plus prioritized fix plan), optimize-ui (Core Web Vitals audit). Backed by a self-contained reference library across 12 domains covering the substance floor (what a surface must have, measured), copy placement and volume, app shells and content layout, fluid and intrinsic sizing, breakpoints versus container queries, zoom and adaptive postures, task flows and journeys, forms and error recovery, navigation and information architecture, states and feedback, WCAG 2.2, OKLCH and APCA color, fluid typography, motion, Tailwind v4, shadcn, component and state architecture, Core Web Vitals, TypeScript 6/7 typing with the TS7 type gate, chart palettes, density and economy, and a 150-plus-tell anti-AI catalogue. Produces severity-tagged findings (CRITICAL/HIGH/MEDIUM/LOW/TASTE) with evidence citations, confidence classes and concrete remediation, against one canonical finding template. Ships a labeled regression corpus (clean and near-miss controls included) with a recall- and precision-gated scoring harness, a deterministic tell scanner with its own corpus selftest, a label contract against the catalogue, a dogfood gate and a reference-integrity check, a CI gate producing GREEN/YELLOW/RED verdict artifacts with an end-to-end selftest, a review ledger for run-over-run delta reports, and an opt-in design hook that checks edited UI files as the agent works. Read-only by defaul
Plugin ui-craft from marketplace ui-craft-public in TheMizeGuy/ui-craft-public. Commands read from the repo's manifest — 3 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
design-uiUse this skill when the user asks to design new UI: a screen, flow, page, component, or full product. Triggers: "design a [thing]", "build me a [screen/page/flow]", "create the UI for", "design the [dashboard/settings/onboarding/landing]", "make this screen look [distinctive/professional/not AI]". D
improve-uiUse this skill when the user wants the full UI treatment: a comprehensive multi-specialist pass covering visual + usability/flow + accessibility + responsive + motion + performance + type safety + anti-AI aesthetics in one review, plus (for improvement asks) a prioritized fix plan. Triggers: "improv
optimize-uiUse this skill when the user asks to optimize UI performance: Core Web Vitals (LCP, INP, CLS), bundle size, rendering performance, runtime smoothness, font loading, image optimization, or framework-specific patterns (React compiler, RSC, Suspense, hydration; SwiftUI/Compose re-render stability). Tri
review-uiUse this skill when the user asks to review any UI for quality: visual defects, alignment, spacing, typography, color, responsiveness, motion, accessibility, task flow, and AI-generated aesthetic tells. Works across web, iOS, Android, desktop, live URLs, and screenshots. Triggers: "review this UI",
ui-accessibility-reviewerRead-only accessibility reviewer for any UI (web, iOS, Android, desktop). Checks semantics, keyboard/focus, contrast, target size, screen reader support, reduced motion, Dynamic Type, VoiceOver, TalkBack, and platform-specific patterns. Returns severity-tagged findings with WCAG citations. Use when
ui-anti-slop-auditorRead-only auditor that verifies UI code does not look AI-generated. Catches AI-default aesthetics (generic gradients, default shadcn, Inter/Roboto, hashtag-purple/teal, Lucide-everywhere, bento-grid-as-default, centered-everything) and returns severity-tagged findings with concrete remediation. The
ui-craft-architectBuilder that designs new UI from scratch — screen, flow, section, component family, or full product. Commits to a distinctive aesthetic POV, maps the user's task flow before any screen exists, generates token systems (OKLCH color, variable fonts, fluid clamp-based type and spacing, spring motion), s
ui-motion-reviewerRead-only motion / animation reviewer for any UI (web, iOS, Android). Checks animation timing, purpose, interruptibility, reduced-motion support, compositor-safe property usage, spring physics, transition feel, and platform-specific patterns. Returns severity-tagged findings with evidence requiremen
ui-perf-engineerRead-only performance + runtime-stability engineer for UI. Web is the primary lane: Core Web Vitals (LCP, INP, CLS), bundle size, font/image loading, rendering, React hydration + server components. It also reviews native rendering performance (SwiftUI body re-evaluation, Compose recomposition, scrol
ui-responsive-reviewerRead-only responsive / cross-viewport reviewer for any UI (web, iOS, Android, desktop). Classifies the sizing strategy (fluid, intrinsic, container-driven, or fixed) before hunting failures, then checks viewport matrices, overflow, clipping, safe areas, zoom reflow, orientation, adaptive postures, a
ui-team-leadDispatched as general-purpose with this body inlined, not through the plugin namespace, because it needs the Agent tool to dispatch the specialists; `skills/improve-ui/SKILL.md` Step 4 is the reference implementation. Orchestrator for the full multi-specialist UI pass. Adaptively dispatches up to 7
ui-typescript-engineerRead-only TypeScript 6/7 strictness + component-typing reviewer for UI code. Reviews prop types, state management, branded primitives, discriminated unions, exhaustiveness, strict-mode, and component API design: the TS quality layer, not the visual layer. Runs the TypeScript 7 type gate + eslint/bio
ui-verifierVerification pass that runs AFTER specialist reviews. Checks evidence sufficiency for every finding, removes false positives, downgrades weak claims, deduplicates cross-dimension findings, enforces the four-class confidence enum, sets the four blocker flags, and emits one canonical verdict token per
ui-visual-reviewerRead-only visual quality, usability, and interaction reviewer for any UI (web, iOS, Android, desktop, design files). Owns two things no other specialist owns: how the surface looks (layout, spacing, alignment, typography, color, hierarchy, component coherence, POV coherence, state completeness, cont
test -f "${CLAUDE_PROJECT_DIR:-.}/.claude/ui-craft/hook.json" || exit 0; node "${CLAUDE_PLUGIN_ROOT}/scripts/design_hook.mjs" post-tool-use || true
test -f "${CLAUDE_PROJECT_DIR:-.}/.claude/ui-craft/hook.json" || exit 0; node "${CLAUDE_PLUGIN_ROOT}/scripts/design_hook.mjs" stop || true
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Editorial diagrams in your brand, as self-contained HTML and SVG. Architecture, flowcharts, sequence, state machines, ER and database schemas, timelines, swimlanes, org charts, Gantt, Wardley maps, user journeys, service blueprints, UML class, lifecycle phase maps, and charts from bar, line and scatter to Sankey, treemap and heatmap, plus many more. Redraws draw.io, Mermaid, Excalidraw and PlantUML files. Exports SVG and PNG.
1 skill · 7 commands
5 skills + 35 modes + Material Passport pipeline. Includes v3.6.7 cross-model audit gate and v3.6.8 generator-evaluator contract.
5 skills · 16 commands · 4 agents · hooks
Perform AI code review on Git diffs — supports workspace changes, branch ranges, and single commits with concurrent per-file analysis, codebase search, and deep context-aware review.
2 commands
daisyUI component library skill
7 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.