Teach your agent anything.

1,251,882 safety-reviewed skills for Claude Code, Cursor, Codex and 60+ more agents.

anthropic
Frontend Design
Guides distinctive, intentional visual design for new UI or existing interfaces, covering aesthetic direction, typography, and avoiding templated defaults.
Audited 158k bundle
anthropic
DOCX
Create, read, edit, and manipulate Word documents (.docx files) with formatting, tables, images, and tracked changes.
158k bundle
anthropic
PPTX
Create, read, edit, and convert PowerPoint (.pptx) files with design guidance and visual QA workflows.
158k bundle
elvarofficial
Teardown
Build an interactive 3D simulator of a real machine or mechanism (an engine, a landing gear, a walking linkage, a pump, a gearbox) as one offline HTML page people can take apart, with live panels driven by the same model that moves the parts, a button for every subsystem, cutaways, camera views and an auto tour that hands control to the visitor; then record it frame by frame to a 60 fps video with a synced procedural score. Use when someone asks how a machine works, wants a "visual explainer", a "simulator", a "cutaway", an "exploded view" or an "interactive 3D lesson", or wants a 3D object shown working rather than described.
265 bundle
anthropic
Skill Creator
Create new skills, modify existing ones, and measure their performance through iterative evaluation and benchmarking.
158k bundle
nvidia
Vss Manage Alerts
Operate the VSS alert pipeline for real-time monitoring, Alert-Bridge subscriptions, Slack notifications, incident queries, and camera onboarding.
Audited 2.2k bundle
nvidia
Jetson LLM Serve
Serve LLMs and VLMs on NVIDIA Jetson devices using vLLM or SGLang with optimized Docker containers and quantization presets.
Audited 2.2k bundle
nvidia
Aiq Research
Runs deep research queries through a reachable NVIDIA AI-Q Blueprint backend, handling health checks, job submission, polling, and report retrieval.
Audited 2.2k bundle
View all
msapps-mobile
Make Plan
Create a work plan using memory context from past sessions. Use when the user says "make a plan", "what should we do next", "plan this out", "give me a roadmap", "help me prioritize", "what's the next step", "plan the sprint", or any request to organize upcoming work. Memory ensures the plan is grounded in what's already done, decided, and known to be broken.
serialforbreakfast
Swift Accessibility Agent
Use this skill whenever the user is asking about accessibility for Apple-platform UI, including SwiftUI, UIKit, AppKit, tvOS, macOS, or visionOS. Trigger on questions about VoiceOver, Accessibility Inspector, Rotor behavior, Dynamic Type, labels, traits, focus, grouping, modal restoration, custom actions, media controls, and semantic regressions. Also use it when the user mentions Apple UI types or APIs such as UILabel, UIButton, UITableView, UICollectionView, UISwitch, NSWindow, NSHostingView, or mixed SwiftUI/UIKit/AppKit stacks and wants accessibility guidance, implementation help, review feedback, or debugging support. The skill selects platform-specific guidance, preserves accessibility semantics, and outputs citations plus verification steps.
bundle
koode
Stew
Reach the user on their desktop through a floating panel (the `stew` CLI) to ask permission for an action, ask a question in AskUserQuestion format, ask for one typed value (a domain, a name, a verification code, a password), tell them something, or tell them you need them to do something only they can do (log in, buy, approve in a web page, a physical action). Use in background jobs and long-running work when the user is likely not watching the terminal, when blocked on a decision or on the user, when a long task finishes or fails, and whenever the user asks to be pinged, notified or asked.
ecorkran
Analysis Understand
Consumes an existing understand-anything knowledge graph and writes squadron planning documents from it. Reads the graph the plugin already produced — it does not analyze the codebase itself and does not run the plugin. Use when the user asks for a codebase comprehension analysis derived from an existing knowledge graph. Does not auto-invoke.
priors-agents
Priors
Pay HTTP 402 (x402) APIs on credit with Priors: when the agent's wallet is short, it borrows the gap in USDG from its Priors credit line, pays, and repays before the due date. Before paying another ERC-8004 agent, read that agent's repayment record. TRIGGERS: "pay this x402 API", "borrow to pay", "repay my loan", "is this agent reliable", "check agent #N's record", HTTP 402, x402, USDG, ERC-8004. Every payment and loan is real money on Robinhood Chain: confirm the URL, the price and the most to borrow with the user first.
henriquebarbosa-cyber
Vulnerability Analysis
Use quando o estudante quiser analisar vulnerailidades, entender riscos de segurança, avaliar falhas em sistemas ou aprender boas práticas de proteção.
bundle
alexk-notable
Self Learn
Capture, triage, and route durable session lessons into canon. Use when the user corrects a mistake Claude made, states a rule or preference that should outlast the current task, names a gotcha that will recur, says "remember this" / "capture this lesson" / "teach", or wants to review, triage, or route pending lessons. Fronts the `self-learn` CLI (git-backed lesson ledger) and the /self-learn:teach and /self-learn:review commands.
bundle
ahmedboin
Jailbreak
Add a new crack or patch project to the Jailbreak repository. Use when reverse engineering a desktop target on any OS, writing the playbook, building a standalone patcher, verifying desktop and secondary media behavior, and publishing under domains/<domain>/<project>/.
bundle
View all

Frequently asked questions

What is an Agent Skill?

An Agent Skill is a reusable instruction file that teaches an AI agent to do one task well, such as reviewing code or extracting PDF tables. It ships as a SKILL.md file: YAML frontmatter plus Markdown instructions. The agent loads it automatically when a request matches, and the same file works across Claude Code, Cursor, Codex, and 60+ other agents.

Where can I download skills for Claude Code and other AI agents?

SkillMD (skillmd.com) is a free, open library of thousands of AI agent skills. Search by task, category, or agent, then install with one CLI command (npx get-skill add owner/name), through the hosted MCP server, or by copying the SKILL.md file directly. Every public skill passes an automated safety review before it is listed.

Do skills work across different agents and IDEs?

Yes. SKILL.md is an open standard, so the same skill works across Claude Code, Claude.ai, Codex, Cursor, Gemini, Antigravity and more. Each tool just reads the file from its skills directory. See skillmd.com/agents for per-agent install paths.

Are skills safe to install?

Treat them like any code you add to your workflow. Every skill on SkillMD shows capability flags (executes scripts, makes network calls, reads secrets, or docs-only), and SkillMD never runs a skill's scripts for you. Review the SKILL.md before installing, especially for unverified authors.

Do AI agent skills cost anything?

No. SkillMD is 100% free: searching, saving, installing, and publishing skills all cost nothing, and skills remain under their authors' original licenses.