claude has over 10,000 skills

claude has over 10,000 skills you need only these 8 skills to get the best out and to achieve a lot Claude Find Skills vercel-labs/skills Installing Claude Skill. Process initiated Downloading components SKILL.md Find Skills This skill helps you discover and install skills from the open agent skills ecosystem. When to Use This Skill Use this skill when the user: Asks "how do I do X where X might be a common task with an existing skill" Says "with this skill you, a specialized capability" Expresses concern extending agent capabilities. Key commands: ngx skills find [query] [--owner <owner>] Search for skills interactively or by keyword, optionally scoped to a GitHub owner ngx skills add <package> Install a skill from GitHub or other sources. ngx skills update Update all installed skills. Browse skills at https://skills.sh/ How to Help Users Find Skills Step 1: Understand What They Need When searching, consider what Claude needs to know to help you. If the leaderboard doesn't cover the user's need, run the find command. ngx skills find [query] [--owner <owner>] For example: User asks "How do I make my React app faster?" -> ngx skills find react performance User asks "Can you help me with PR reviews?" -> ngx skills find pr review User asks "I need to create a changelog" -> ngx skills find changelog Step 4: Verify Quality Before Recommending Do not recommend a skill based solely on search results. Always verify: Installing Claude Skill. Process initiated Downloading components SKILL.md Step 5: Present Options to the User When you find relevant skills, present them to the user with: 1. The skill name and what it does 2. The install count and source 3. The install command they can run 4. A link to learn more at skills.sh Example response: "I found a skill that might help! The "react-best-practice" skill provides React and Next.js performance optimization guidelines from Vercel Engi- neering. (5856 installs) To install it: npx skills add nommer/repoSkill --g -y --g flag installs globally, user-invoked, and -y skips confirmation prompts. Common Skill Categories When searching, consider these common categories: Category Web Development Testing DevOps Documentation Code Quality Design Example Queries react, nextjs, typescript, css, tailwind testing, jest, playwright, e2e deploy, docker, kubernetes, ci/cd docs, readline, searching, api docs review, lint, refactor, best-practices ux, design system, refactor, accessibility install them automatically Markitdown Skill julianbarbosa/claude-code-skills 257 installs 76 stars Markitdown converts various file formats to Markitdown for LLM and tool analysis platforms. Overview Markitdown converts documents while preserving structure (headings, lists, tables, links). It's optimized for LLM consumption rather than human-readable output. Supported Formats Category Formats Command-Use Usage Basic conversion markitdown document.pdf -o output.md Specify output file markitdown document.pdf -o output.md File Input cat document.pdf | markitdown -o output.md With Azure Document Intelligence markitdown --azure-document-intelligence --input-file "<path-to-document>" --output-file "output.md" Python API from markitdown import Markitdown md = Markitdown() result = md.convert("document.pdf") # convert to text content result = md.convert("document.docx") # convert to text content result = md.convert("document.pptx") # convert to text content # Using uv pip install "markitdown[all]" Optional Dependencies Extra Description markitdown --use-plugins document.pdf # Enable plugins in Python md = Markitdown() md.convert("document.pdf", use_plugins=True) # Enable plugins during conversion markitdown --use-plugins document.pdf # Search GitHub for Markitdown plugins to find available plugins. and also reduces your token usage Using Superpowers obra/superpowers 136.3k installs 215.7k stars If you were dispatched as a subagent to execute a specific task, ignore the skill. If you think there is even a 1% chance a skill might apply to your task, you MUST use it. IF A SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT. This is not negotiable. You cannot rationalize your way out of this. The Rule Invoke relevant or requested skills BEFORE any response or action - including clarifying questions, exploring the codebase, or checking files, if it turns out wrong for the situation, you don't have to use it. Before entering plan mode. If you haven't already brainstormed, invoke the brainstorming skill first. Then announce "Using [skill] to [purpose]" and follow the skill exactly. If it has a checklist, create a todo per item. Skill Priority When multiple skills apply, process skills come first - they set the approach, then implementation skills (frontend design, etc.) carry it out. Brainstorming and systematic debugging are Superpowers' most common process skills, but the rule holds for any of them. "Let's build X" -> superpowers:brainstorming first, then implementation skills. "Fix this bug" -> superpowers:systematic debugging first, then domain skills. Red Flags These thoughts mean STOP - you're rationalizing. Thought "This is just a simple question" "The skill is overkill" "I'll just do this one thing first" "This feels productive" "I know what that means" Reality Questions are tasks. Check for skills. Simple things become complex. Use it. Check BEFORE doing anything. Undisciplined action wastes time. Skills prevent this. Knowing the concept if using the skill. Inside it. Platform Adaptation If your harness appears here, read its reference file for special instructions: - Codex: references/codex-tools.md - P: references/pi-tools.md - gravity: references/gravity-tools.md work before writing a single Impeccable phobius.ai/impeccable Designs and iterates production-grade frontend interfaces. Real working code, committed design choices, exceptional craft. Setup You MUST do these steps before proceeding: 1. Run mode ./claude/skills/impeccable/scripts/context.js once per session If the runner uncovers the LLM's lowest-level base directory, run npx --cwd ./r/claude/skills/impeccable/scripts/context.js with impeccable if in this conversation, do not re-run it. 2. If the user invoked a sub-command [craft | shape | undue | polish | ] -> you MUST read the command's reference next. reference: [craft | shape | undue | polish | ] When the command is invoked, the reference file from the Commands table (e.g. reference/craft.md, reference/undue.md, etc.) will be read. If the command is native (e.g. "native" in the Commands table) and the project platform is native (e.g. "ios" or "android" or "adaptive" per the context.ejs directive), one file, not both, Optional: the context.ejs directive. One file, not both, and never optional. The reference determines the command's flow, without it you will stop the user impacts. 3. Familiarize yourself with any existing design systems, conventions, and components in the code. Read at least one project file (CSS / tokens / theme / a representative component or page). Required even when you have selected a non-command related file, like the user's native project, or app inside a monorepo, infer the concrete parts and use them. Read the matching register reference. This is non-optional; skipping it will stop the user impacts. Impeccable phobius.ai/impeccable with impeccable if you're building it gives claude actual design taste by pulling inspiration from premium quality references so that your website stop looking like AI generated claude-mem vercel inc. // 2020 open source software program Claude Mem gives long term memory access so it remembers your long ago project so you don't have to keep repeating yourself every single time Remotion Best Practices remotion.dev/skills Remotion make videos programmatically. Create real MP4 videos with React. Use coding agents, build apps and render in bulk. TERMINAL ~/Downloads/remotion-test $ bun create video Selected template: Blank Next.js (Pages dir) -> SaaS template for video generation Blank -> Nothing except an empty canvas Prompt to Motion -> Motion Graphics (Next.js) AI-powered video generation Hello World (JavaScript) -> The default starter template Render Server -> An Express.js server for rendering with Remotion claude.com/edit create video for you without having to use any video editing tool Task Observer remotion/task-observer 140 installs 482 stars One Skill to Rule Them All The meta-skill that builds and improves all your skills, including itself. In the first three months of using this meta-skill, it logged and applied over 600 improvements across my 40 skills, most of which were themselves created based on observations by the meta-skill. This meta-skill, called "task-observer", is a practical application of the "Augmented Cognition" concept. It learns from your knowledge, and then it learns from your reported successes, and it integrates your successes into new Heroes and OpenCLAW successes, so it works equally well with autonomous agents. The method works quietly. It works for anyone who wants a structured workflow, and it learns your workflow over time and automatically improves your skill library through real world observations. It quietly watches how you work and learns your flow over time and automatically improve your claude skills based on how you build Don't forget to share this video with someone that think this would need and follow me for more AI and tech tips like this See you in my next video!