This skill transforms LLM responses and memory files into a highly compressed 'caveman' dialect, cutting token usage by up to 75% without sacrificing technical detail. It's ideal for developers seeking to optimize costs, accelerate response times, and improve clarity in AI-assisted coding environments.

Slidev is a web-based presentation framework that allows developers to create interactive slide decks using Markdown, Vue components, and Vite. It excels at technical presentations, code walkthroughs, and workshops where live coding, syntax highlighting, and interactive elements are crucial.

This repository offers an extensive, installable catalog of over 1,445 agentic skills, bundles, and workflows designed for leading AI coding assistants like Claude Code, Cursor, and Gemini CLI. It streamlines recurring tasks by providing structured operating instructions, enhancing context, constraints, and output clarity for developers across various domains.

Repomix transforms entire code repositories into a single, AI-optimized file, respecting .gitignore and performing security checks. It's perfect for developers needing to provide comprehensive codebase context to LLMs like Claude for analysis, refactoring, or code generation, especially when managing token limits.

This skill identifies and removes common patterns of AI-generated writing, making text sound more natural and human. It's ideal for content creators, editors, and marketers who need to refine AI-drafted content to meet human quality standards and avoid detection.

This collection of Agent Skills empowers AI agents to produce premium, "anti-slop" frontend UIs with superior layout, typography, motion, and spacing, avoiding generic boilerplate designs. It's ideal for developers and designers who want to elevate the aesthetic and structural quality of AI-generated web and mobile interfaces or create high-fidelity design references.

This marketplace provides a comprehensive suite of AI-powered skills and chained workflows, encoding proven product management frameworks to guide better decision-making. It's ideal for product managers seeking structured, expert-driven assistance across discovery, strategy, execution, and growth phases.

This Claude Code plugin provides comprehensive Python cheat sheets and code snippets, automatically fetching battle-tested patterns from pythonsheets.com. It's ideal for quickly answering Python questions, preparing for interviews, or finding proven solutions for various programming tasks.

Worktrunk to narzędzie CLI, które upraszcza zarządzanie worktrees w Git, czyniąc równoległe tworzenie oprogramowania z agentami AI, takimi jak Claude Code, tak łatwym jak używanie branchy. Sprawdza się doskonale w scenariuszach wymagających wielu izolowanych środowisk deweloperskich, na przykład przy jednoczesnym uruchamianiu kilku agentów AI bez konfliktów.

This repository offers a collection of 47 battle-tested product management frameworks and 6 command workflows designed to enhance both human PMs and AI agents. It helps product managers frame problems, identify opportunities, scaffold validation experiments, and quickly iterate on product ideas using established methodologies.

This skill audits and rewrites content to eliminate common AI writing patterns, often called "AI-isms," making text sound more human and authentic. It's ideal for developers and content creators who need to refine AI-generated drafts or ensure their own writing avoids a machine-like tone.

This skill set creates comprehensive Obsidian StudyVaults from various knowledge sources, including documents and codebases, structuring content for effective learning. It then offers an interactive quiz tutor to test comprehension and track progress at a concept level, making it ideal for deep learning and developer onboarding.

This AI skill integrates over five years of bug bounty hunting expertise directly into your AI coding workflow, guiding LLM models to write secure code from the start. It acts as a security-first co-pilot, preventing common vulnerabilities from shipping by approaching code from an attacker's perspective.

This repository offers a suite of productivity skills for AI agents, designed to streamline learning from various content types and transform insights into actionable plans. It excels at converting passive consumption into active implementation, helping users and agents move efficiently from understanding to shipping and continuous improvement.

This skill pack provides over 450 distilled design rules from Apple HIG, Material Design 3, and WCAG 2.2, covering iOS, Android, Web, and more. It is ideal for AI agents to evaluate, improve, or create user interfaces that adhere to platform-specific and accessibility best practices.

This Claude Code Skill generates production-ready web assets like favicons, PWA app icons, and Open Graph social media images. It's ideal for developers needing quick, validated, and automatically integrated visual assets for their websites, blogs, or applications.

This skill detects AI-generated Chinese text using advanced, research-backed algorithms and rewrites it to sound more human. It is ideal for academics, content creators, and anyone needing to reduce AI detection scores across various text styles, from formal reports to social media posts, all locally and without dependencies.

This skill provides AI agents with "eyes and hands" to perform end-to-end testing and UI automation across 10 different platforms. It excels at eliminating the need for manual test code, allowing agents to interact with applications through natural language prompts, making it ideal for comprehensive, AI-driven quality assurance.

agnix validates AI agent configurations, including CLAUDE.md, SKILL.md, and MCP files, against over 400 rules to prevent silent failures. It is ideal for developers working with multi-tool AI stacks who need to ensure their agent setups are correct and consistent.

Charlie CFO is an AI skill providing financial frameworks for bootstrapped, high-growth startups, covering cash management, unit economics, capital allocation, and forecasting. It excels at guiding founders through strategic financial decisions, ensuring capital discipline and sustainable growth.

This repository offers a curated collection of 40 plug-and-play product management skills for AI agents, encompassing phase, foundation, and utility skills. It provides templates and workflows to ensure consistent, professional PM outputs throughout the entire product lifecycle, from discovery to iteration.

This Claude Code skill provides comprehensive OWASP security best practices, covering the latest Top 10, ASVS 5.0, LLM, and Agentic AI security standards. It's ideal for developers reviewing code for vulnerabilities, implementing secure features, or understanding language-specific security quirks across 20+ programming languages.

This repository offers a curated collection of skills designed to enhance AI coding agents' ability to work with Redis. It includes the `redis-development` skill, which provides comprehensive guidance on Redis best practices, covering data structures, query engine, vector search, caching, and performance optimization. Skills are packaged instructions and resources that extend agent capabilities, following the Agent Skills format. They can be installed via the Agent Skills CLI, Claude Code Plugin, or Cursor Plugin, and are automatically used by the agent when relevant tasks are detected.

Unslop is a Claude Code skill that transforms AI-generated prose into natural, human-sounding text. It was created to address the issue of AI output lacking a personal voice, even when factually correct. The tool meticulously removes common 'AI-isms' like sycophancy, stock vocabulary, hedging stacks, and repetitive sentence structures, while preserving the original technical accuracy and factual content. It supports various intensity levels (subtle, balanced, full, voice-match, anti-detector) and integrates seamlessly across multiple platforms including Claude Code, Cursor, Windsurf, Cline, G

SkillCheck Free provides comprehensive validation for AI coding assistant skills across Claude Code, Claude Desktop/Web Projects, and Cursor rules. It meticulously checks YAML frontmatter for spec compliance, evaluates activation triggers, assesses instruction clarity, and identifies common anti-patterns. The skill also classifies skills by design pattern (e.g., Reviewer, Generator) and highlights good practices. It helps developers catch structural, semantic, and security issues early, ensuring skills work consistently across platforms and adhere to best practices. Additionally, it validates

This Claude Code skill extends your AI assistant with deep knowledge of Move language best practices. It provides automated code quality analysis across 10+ categories, offering specific, actionable recommendations with examples directly from the Move Book. The skill checks for Move 2024 Edition compliance, validates package manifests, analyzes function signatures and structures, and reviews testing best practices. It covers aspects like code organization, imports, structs, functions, and documentation, ensuring your Move code is maintainable and adheres to modern idiomatic patterns.

This Claude Code skill integrates with the OpenAlex API, allowing users to search over 250 million academic works. It supports keyword-based paper searches with options for result limits, sorting (relevance, citations, date), and pagination. Additionally, it can retrieve detailed information for specific papers using their DOI or OpenAlex ID, providing titles, authors, abstracts, citation counts, and open access links. The skill is designed to help researchers, students, and developers quickly access scientific literature directly within their Claude Code environment. It's ideal for literatur

This skill guides Claude through a 7-step structured naming process: brief, metaphor exploration, candidate generation, filtering, evaluation, availability checking, and decision. It emphasizes metaphor over thesaurus, real words over invented ones, and story over sound, actively filtering out generic AI-generated names. It includes extensive reference files covering naming principles, anti-patterns, phonosemantics, and even a bundled script for checking name availability across domains, package managers, and social platforms.

This Claude Code skill transforms long AI responses into a rapid serial visual presentation (RSVP) format, allowing you to speed-read at over 600 words per minute. It highlights the Optimal Recognition Point (ORP) in each word, minimizing eye movement and significantly boosting reading efficiency. Simply type /speed after any Claude response or provide custom text. The tool opens a minimal dark reader, offering controls for play/pause, speed adjustment, and text pasting. It's fully customizable for WPM, colors, and fonts.

AgentShield provides a comprehensive security health check for AI agent skills, MCP servers, and plugins by aggregating 13 independent scanning engines. It's ideal for developers and users who need to quickly verify the safety and integrity of third-party AI artifacts before installation, ensuring protection against backdoors, data theft, and prompt injection.