This repository offers Anthropic's official implementation of skills for Claude, providing a diverse collection of examples ranging from creative applications to technical tasks and enterprise workflows. It serves as an excellent resource for developers to understand, get inspired by, and extend Claude's capabilities through custom skills.

This AI skill provides comprehensive design intelligence for building professional UI/UX across multiple platforms and frameworks. It features an AI-powered Design System Generator that analyzes project requirements to create tailored design systems, including patterns, styles, colors, and typography, making it ideal for rapid prototyping and ensuring visual consistency.

Impeccable is an AI skill that elevates frontend design by providing 23 specialized commands and a curated set of anti-pattern rules. It helps developers and designers move beyond generic AI output, ensuring high-quality, production-grade interfaces with exceptional craft and adherence to best practices in typography, color, motion, and UX writing.

Slint is an open-source declarative GUI toolkit for building native user interfaces across embedded, desktop, and mobile platforms using Rust, C++, JavaScript, or Python. It excels in creating responsive, lightweight, and native-feeling applications with a separate UI markup language, making it ideal for projects requiring efficient cross-platform UI development and AI-assisted debugging.

Frontend Slides is a Claude Code skill that generates beautiful, animation-rich HTML presentations from scratch or by converting existing PowerPoint files. It's ideal for non-designers who want to create visually distinctive web presentations without needing CSS or JavaScript knowledge, offering visual style previews to guide aesthetic choices.

Tambo AI is an open-source React toolkit for building agents that render dynamic UI components. It excels when you need to integrate AI-powered interfaces directly into your application, allowing agents to respond to user input by generating and updating React components.

AntV Infographic to deklaratywny silnik wizualizacji, który szybko renderuje wysokiej jakości infografiki na podstawie starannie zaprojektowanej składni. Doskonale sprawdza się w przekształcaniu złożonych informacji w efektywne, angażujące wizualnie historie danych, szczególnie w połączeniu z agentami AI do automatycznego tworzenia treści.

This skill helps Claude Code build professional, polished interfaces by remembering design decisions across sessions and enforcing systematic consistency. It is ideal for crafting dashboards, applications, tools, and admin panels, ensuring every component follows established principles.

This repository offers a suite of UI-focused skills designed to assist design engineers in maintaining high-quality user interfaces. It includes tools for enforcing design baselines, auditing accessibility, optimizing animation performance, and managing SEO metadata. These skills are invaluable for ensuring consistency, compliance, and optimal user experience across web projects.

This Claude Code skill enables the creation of polished, professional Reveal.js HTML presentations using natural language. It excels at generating visually appealing slides with themes, multi-column layouts, and data visualizations, ideal for quickly producing engaging content without manual HTML coding.

This skill generates and edits images using OpenAI's GPT Image 2 model, offering superior text rendering and a "thinking mode" for complex compositions like infographics or diagrams. It's ideal for creating social media content, transforming photos into artistic styles, or any image generation where high-quality text in visuals is crucial.

This Model Context Protocol (MCP) server integrates the SVGMaker API, allowing AI assistants to generate, edit, and convert SVG images. Users can create SVGs from text descriptions, modify existing vector graphics using natural language, and convert raster images to scalable SVGs. It offers real-time progress updates, secure file operations, and full TypeScript support. The server is designed for seamless integration with various LLM environments like Claude Desktop, Cursor, VS Code, WindSurf, and Zed, providing specific configuration examples for each.

This skill empowers agents to design high-quality SwiftUI interfaces that look intentionally crafted, not generic or AI-generated. It enforces 6 Anti-AI Slop Iron Rules, provides a Design Direction Advisor for vague requirements, integrates brand assets via a strict protocol, and ensures quality with a 5-Dimension Review. Developers can generate compilable SwiftUI code, design system tokens, and layout patterns. It's agent-agnostic, supporting Claude Code, Cursor, Codex, and OpenCode, aiming to produce a signature iOS interface in minutes, complete with visual anchor descriptions and actionab

This skill offers a structured UI/UX consultation across three modes: "design" (co-create a project-specific design system, outputting `design-spec.md`), "review" (critique existing UI, providing prioritized P0/P1/P2 fixes), and "guide" (generate 'do/don't' rules for specific page types). It acts as a patient interviewer, prioritizing your product's brand, references, and constraints. It clearly separates universal UX Hard Rules from project-selectable Style Lenses (e.g., modern-minimal, brutal, playful), ensuring tailored guidance. The skill scans existing code (Tailwind, UI frameworks, CSS)

This AI skill transforms natural language prompts into hand-drawn diagrams in Excalidraw's sketch style. It automatically infers the diagram type, lays out elements, and provides a hosted edit URL, an animated SVG that draws itself stroke by stroke, and static PNG exports. It streamlines diagram creation by removing the need for manual drawing and layout, offering a clean workspace by saving temporary files. It's ideal for quickly visualizing ideas, systems, and workflows without opening a dedicated app.

This repository provides a curated, local marketplace for Claude Code skills, enabling easy installation and management of custom skills. It supports both global Claude Code plugins and repo-local Codex skills, offering helper scripts for seamless integration. The marketplace includes a diverse collection of skills for development workflows, UI design, Git operations, terminal automation, and advanced AI reasoning.

Nakkas is an MCP server that transforms AI prompts into rich, animated SVG graphics. It operates on a principle of "one rendering engine, AI decides everything," using a pure declarative SVG approach with CSS @keyframes and SMIL animations, without JavaScript. The tool provides an AI-native schema, where every field has `.describe()` annotations, allowing models to understand and generate precise JSON configurations. It runs locally with zero external dependencies, offering `render_svg`, `preview`, and `save` tools for a streamlined workflow: render, visually inspect, iterate, and then save t

This Model Context Protocol (MCP) server enables AI image generation and editing capabilities, powered by Google's Nano Banana model and accessed through the AceDataCloud API. It integrates seamlessly with various MCP-compatible clients like Claude, VS Code, JetBrains IDEs, and more. Key features include creating high-quality images from text prompts, modifying existing images, combining multiple images, virtual try-on, and product placement. Users can choose between a managed hosted server (with OAuth for Claude.ai) or self-hosting via PyPI or Docker.