Ui Toolkit Web logo

Ui Toolkit Web

Organization
zoom
ui-toolkit-web

Zoom Video SDK UI Toolkit for Web. Use when building prebuilt or low-code Zoom Video SDK browser UI with @zoom/videosdk-ui-toolkit, composite UI, individual components, session lifecycle, React/Vue/Angular/Next.js/vanilla integration, featuresOptions, JWT auth, CSS/CDN install, troubleshooting, or deciding between UI Toolkit and raw Video SDK.

Overview

Publisherzoom
Repositoryskills
Skill nameui-toolkit-web
Stars
78
Forks
16
Bundled files
3
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by zoom on GitHub. Read the source before you install it.

Installation

Install the Ui Toolkit Web AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/zoom/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/skills/ui-toolkit .claude/skills/ui-toolkit-web
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ui Toolkit Web in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Ui Toolkit Web on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Ui Toolkit Web is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Zoom Video SDK UI Toolkit

Pre-built video conferencing UI powered by Zoom Video SDK. Drop-in solution for web applications.

Official Documentation: https://developers.zoom.us/docs/video-sdk/web/ui-toolkit/ API Reference: https://marketplacefront.zoom.us/sdk/uitoolkit/web/ NPM Package: https://www.npmjs.com/package/@zoom/videosdk-ui-toolkit Live Demo: https://sdk.zoom.com/videosdk-uitoolkit

Quick Links

New to UI Toolkit? Follow this path:

  1. Quick Start - Get running in 5 minutes (see below)
  2. JWT Authentication - Server-side token generation (required)
  3. Composite vs Components - Choose your approach
  4. Framework Integration - React, Vue, Angular, Next.js patterns
  5. Integrated Index - see the section below in this file

Having issues?

  • Session not joining → Check JWT Authentication (most common issue)
  • React 18 peer dependency error → See Installation section
  • CSS not loading → See Troubleshooting
  • Components not showing → Check Component Lifecycle
  • Start with preflight checks → 5-Minute Runbook

Overview

The Zoom Video SDK UI Toolkit is a pre-built video UI library that renders complete video conferencing experiences with minimal code. Unlike the raw Video SDK, the UI Toolkit provides:

  • Ready-to-use UI - Professional video interface out of the box
  • Zero UI code - No need to build video layouts, controls, or participant management
  • Framework agnostic - Works with React, Vue, Angular, Next.js, vanilla JS
  • Highly customizable - Choose which features to enable, customize themes
  • Built-in features - Chat, screen share, settings, virtual backgrounds included

When to use UI Toolkit:

  • You want a complete video solution quickly
  • You need Zoom-like UI consistency
  • You don't want to build custom video UI
  • You need standard features (chat, share, participants)

When to use raw Video SDK instead:

  • You need complete custom UI control
  • You're building a non-standard video experience
  • You need access to raw video/audio data
  • You want to build your own rendering pipeline

Installation

bash
npm install @zoom/videosdk-ui-toolkit jsrsasign
npm install -D @types/jsrsasign

Current package: @zoom/videosdk-ui-toolkit@2.4.5-1 is the latest npm release verified on 2026-07-10. Check npm before pinning.

Peer dependencies for 2.4.5-1: @zoom/videosdk@^2.4.5, React 18, React DOM 18, @reduxjs/toolkit@^2, react-redux@^9, and react-draggable@^4.

Upgrade guardrail: Use 2.3.15-1 or newer for WebRTC video because Zoom flags that version floor for future Chrome compatibility.

Need Video SDK credentials first? Use Marketplace app management for creating or validating the Video SDK Marketplace app and credential shape, then return here to generate the UI Toolkit videoSDKJWT.

Quick Start

Basic Usage (Vanilla JS)

javascript
import uitoolkit from "@zoom/videosdk-ui-toolkit";
import "@zoom/videosdk-ui-toolkit/dist/videosdk-ui-toolkit.css";

const container = document.getElementById("sessionContainer");

const config = {
  videoSDKJWT: "your_jwt_token",
  sessionName: "my-session",
  userName: "John Doe",
  sessionPasscode: "",
  featuresOptions: {
    video: { enable: true },
    audio: { enable: true },
    share: { enable: true },
    chat: { enable: true, enableEmoji: true },
    users: { enable: true },
    settings: { enable: true },
    leave: { enable: true },
  },
};

async function startSession() {
  await uitoolkit.joinSession(container, config);

  uitoolkit.onSessionJoined(() => {
    console.log("Session joined");
  });

  uitoolkit.onSessionClosed(() => {
    console.log("Session closed");
  });
}

void startSession();

Next.js / React Integration

typescript
'use client';

import { useEffect, useRef } from 'react';

export default function VideoSession({ jwt, sessionName, userName }) {
  const containerRef = useRef<HTMLDivElement>(null);
  const uitoolkitRef = useRef<any>(null);

  useEffect(() => {
    let isMounted = true;

    const init = async () => {
      const uitoolkitModule = await import('@zoom/videosdk-ui-toolkit');
      const uitoolkit = uitoolkitModule.default;
      uitoolkitRef.current = uitoolkit;
      
      // If TypeScript complains about CSS imports, configure your app to allow them
      // (for example via a global `declare module \"*.css\";`), or import the CSS from
      // a global entrypoint (Next.js layout/_app) instead of inlining here.
      await import('@zoom/videosdk-ui-toolkit/dist/videosdk-ui-toolkit.css');

      if (!isMounted || !containerRef.current) return;

      const config: any = {
        videoSDKJWT: jwt,
        sessionName: sessionName,
        userName: userName,
        sessionPasscode: '',
        featuresOptions: {
          video: { enable: true },
          audio: { enable: true },
          share: { enable: true },
          chat: { enable: true, enableEmoji: true },
          users: { enable: true },
          settings: { enable: true },
          leave: { enable: true },
        },
      };

      await uitoolkit.joinSession(containerRef.current, config);
      uitoolkit.onSessionJoined(() => console.log('Joined'));
      uitoolkit.onSessionClosed(() => console.log('Closed'));
    };

    init();

    return () => {
      isMounted = false;
      if (uitoolkitRef.current && containerRef.current) {
        void uitoolkitRef.current
          .closeSession(containerRef.current)
          .then(() => uitoolkitRef.current.destroy())
          .catch(() => {});
      }
    };
  }, [jwt, sessionName, userName]);

  return <div ref={containerRef} style={{ width: '100%', height: '100vh' }} />;
}

Available Features

FeatureDescription
previewPre-join device, microphone, speaker, name, and background flow
videoEnable video layout and send/receive video
audioAudio toolbar controls, device handling, noise suppression options
secondaryAudioEnable secondary audio support
shareScreen sharing controls
chatIn-session chat, including emoji option
usersParticipant list
settingsDevice configuration, virtual background, quality statistics
recordingCloud recording button; paid plan required
phoneJoin-by-phone options; paid plan required
inviteInvite link/options
themeTheme picker and default theme
viewModeView-mode switcher and defaults
feedbackEnd-of-session experience feedback
troubleshootTroubleshooting tab powered by Zoom Probe SDK
captionIn-session translation/caption UI; paid plan required
playbackMedia file playback options
subsessionSubsession/breakout-room button
leaveLeave/end session button
virtualBackgroundBuilt-in and custom virtual background options
footerFooter button area
headerSession header area
screenshotVideo/share screenshot feature toggles

Use featuresOptions for new code. The older top-level features, options, and virtualBackground fields are deprecated in the current TypeScript definitions.

The GitHub source also exposes advanced feature options that are not all listed in the README feature table. Treat these as source-validated advanced surfaces and verify against the installed package types before shipping:

Advanced optionUse
video.ptzEnable PTZ camera support for PTZ-capable cameras
audio.muteEntryJoin session audio muted
share.controlsEnable supported browser display-media controls
share.displaySurfaceEnable supported browser surface selection hints
share.hideShareAudioOptionHide share-computer-audio option where supported
share.optimizedForSharedVideoPrefer frame rate for shared video content
livestreamLive stream UI support
crcRoom system / CRC invite support
whiteboardWhiteboard UI, export, and viewer-export flags
rawDataSource-level raw data configuration surface
realTimeMediaStreamsSource-level RTMS configuration surface
cameraShareSource-level camera share entry point

If TypeScript does not expose one of these options in the installed package, prefer the typed README/d.ts surface for production code or gate the source-only field behind a version check and local type augmentation.

Troubleshooting

JWT Token Generation (Server-Side)

Required: Generate JWT tokens on your server, never expose SDK secret client-side.

Node.js / Next.js API Route

typescript
import { NextRequest, NextResponse } from 'next/server';
import { KJUR } from 'jsrsasign';

const ZOOM_VIDEO_SDK_KEY = process.env.ZOOM_VIDEO_SDK_KEY;
const ZOOM_VIDEO_SDK_SECRET = process.env.ZOOM_VIDEO_SDK_SECRET;

export async function POST(request: NextRequest) {
  const { sessionName, role, userName } = await request.json();

  if (!sessionName || role === undefined) {
    return NextResponse.json({ error: 'Missing params' }, { status: 400 });
  }

  const iat = Math.floor(Date.now() / 1000);
  const exp = iat + 60 * 60 * 2; // 2 hours

  const oHeader = { alg: 'HS256', typ: 'JWT' };
  const oPayload = {
    app_key: ZOOM_VIDEO_SDK_KEY,
    role_type: role, // 0 = participant, 1 = host
    tpc: sessionName,
    version: 1,
    iat,
    exp,
    user_identity: userName || 'User',
  };

  const signature = KJUR.jws.JWS.sign(
    'HS256',
    JSON.stringify(oHeader),
    JSON.stringify(oPayload),
    ZOOM_VIDEO_SDK_SECRET
  );

  return NextResponse.json({ signature });
}

JWT Payload Fields

FieldRequiredDescription
app_keyYesYour Video SDK Key
role_typeYes0 = participant, 1 = host
tpcYesSession/topic name
versionYesAlways 1
iatYesIssued at (Unix timestamp)
expYesExpiration (Unix timestamp)
user_identityNoUser identifier

API Reference

Core Methods

javascript
uitoolkit.openPreview(container, config, { onClickJoin });
uitoolkit.closePreview(container);
await uitoolkit.joinSession(container, config);
await uitoolkit.leaveSession();
await uitoolkit.closeSession(container);
await uitoolkit.destroy();
const migrated = uitoolkit.migrateConfig(oldConfig);

Event Listeners

javascript
uitoolkit.onSessionJoined(callback);
uitoolkit.onSessionClosed(callback);
uitoolkit.onSessionDestroyed(callback);
uitoolkit.onViewTypeChange(callback);
uitoolkit.onLanguageChange(callback);
uitoolkit.on("user-added", callback);
uitoolkit.offSessionJoined(callback);
uitoolkit.offSessionClosed(callback);
uitoolkit.offSessionDestroyed(callback);
uitoolkit.offViewTypeChange(callback);
uitoolkit.offLanguageChange(callback);
uitoolkit.off("user-added", callback);

Component Methods

javascript
uitoolkit.showChatComponent(container);
uitoolkit.hideChatComponent(container);
uitoolkit.showUsersComponent(container);
uitoolkit.hideUsersComponent(container);
uitoolkit.showControlsComponent(container);
uitoolkit.hideControlsComponent(container);
uitoolkit.showSettingsComponent(container);
uitoolkit.hideSettingsComponent(container);
uitoolkit.hideAllComponents();

The repo source/runtime bundle also includes broadcast viewer helpers:

javascript
uitoolkit.showBroadcastViewerComponent(container, options);
uitoolkit.hideBroadcastViewerComponent(container);

These may not be present in every published .d.ts; verify the installed package before relying on TypeScript autocomplete.

Utility and Advanced Methods

javascript
uitoolkit.getSessionInfo();
uitoolkit.getCurrentUserInfo();
uitoolkit.getAllUser();
uitoolkit.getClient(); // available when debug mode is true
uitoolkit.version();
uitoolkit.changeViewType("gallery");
uitoolkit.mirrorVideo(true);
uitoolkit.isSupportCustomLayout();
uitoolkit.subscribeAudioStatisticData({ encode: true, decode: true });
uitoolkit.subscribeVideoStatisticData({ encode: true, decode: true, detailed: true });
uitoolkit.subscribeShareStatisticData({ encode: true, decode: true });
uitoolkit.i18n.setLanguage("en-US");
uitoolkit.i18n.setLanguageList(["en-US", "zh-CN"]);
uitoolkit.i18n.getLanguage();
uitoolkit.i18n.getLanguageList();
uitoolkit.i18n.hasLanguageResources("ja-JP");
uitoolkit.ptz.isBrowserSupported();
uitoolkit.ptz.getCapability();
uitoolkit.ptz.getFarEndCapability(userId);
uitoolkit.ptz.requestControl(userId);
uitoolkit.ptz.approveControl(userId);
uitoolkit.ptz.declineControl(userId);
uitoolkit.ptz.giveUpControl(userId);
uitoolkit.ptz.control({ cmd: 1, userId, range: 50 });

CDN Usage (No Build Step)

Example version: 2.4.5-1. Replace {VERSION} with the package version you have validated.

html
<link rel="stylesheet" href="https://source.zoom.us/uitoolkit/{VERSION}/videosdk-ui-toolkit.css" />
<script src="https://source.zoom.us/uitoolkit/{VERSION}/videosdk-ui-toolkit.min.umd.js"></script>

<div id="sessionContainer"></div>

<script>
  const uitoolkit = window.UIToolkit;
  
  uitoolkit.joinSession(document.getElementById('sessionContainer'), {
    videoSDKJWT: 'your_jwt',
    sessionName: 'my-session',
    userName: 'User',
    featuresOptions: {
      video: { enable: true },
      audio: { enable: true },
      chat: { enable: true, enableEmoji: true },
      leave: { enable: true }
    }
  });
</script>

Next.js with basePath

When deploying Next.js under a subpath:

typescript
// next.config.ts
const nextConfig = {
  basePath: "/your-app-path",
  assetPrefix: "/your-app-path",
};

Fetch API routes with full path:

typescript
fetch('/your-app-path/api/token', { ... })

Prerequisites

  1. Zoom Video SDK credentials from Zoom Marketplace
  2. React version compatible with your installed UI Toolkit package (check peer deps; React 18 is common)
  3. Server-side JWT generation (never expose SDK secret)
  4. Modern browser with WebRTC support

Browser Support

BrowserVersion
Chrome78+
Firefox76+
Safari14.1+
Edge79+

Common Issues

IssueSolution
peer react@"^18.0.0" errorUse the React version required by the installed UI Toolkit package (check peer deps; React 18 is common)
CSS import TypeScript errorConfigure TS/CSS handling (prefer a global *.css module declaration); avoid @ts-ignore except in throwaway demos
Config type errorType config as any
API returns HTML not JSONCheck basePath in fetch URL

Resources


Integrated Index

This section was migrated from SKILL.md.

Complete navigation for all UI Toolkit documentation.

📚 Start Here

New to the UI Toolkit? Follow this learning path:

  1. SKILL.md - Main overview and quick start
  2. 5-Minute Runbook - Preflight checks before deep debugging
  3. Quick Start Guide - Working code in 5 minutes (see skill.md)
  4. JWT Authentication - Server-side token generation (see skill.md)
  5. Choose Your Mode - Composite vs Components (see skill.md)

🎯 Core Concepts

Understanding how UI Toolkit works:

  • Composite vs Components - Two ways to use UI Toolkit (see skill.md)
  • UI Toolkit Architecture - How it wraps Video SDK internally
  • Feature Configuration - Understanding featuresOptions structure
  • Session Lifecycle - Join → Active → Leave/Close → Destroy flow

📖 Complete Guides

Getting Started

  • Installation - NPM install and React 18 setup (see skill.md)
  • Quick Start - Composite - Full UI in one container (see skill.md)
  • Quick Start - Components - Individual UI pieces (see skill.md)
  • JWT Authentication - Server-side token generation (see skill.md)

Framework Integration

  • React Integration - Hooks, useEffect patterns (see skill.md)
  • Vue.js Integration - Composition API and Options API (see skill.md)
  • Angular Integration - Component lifecycle (see skill.md)
  • Next.js Integration - App Router, Server Components (see skill.md)
  • Vanilla JavaScript - No framework usage (see skill.md)

Advanced Topics

  • Component Lifecycle - Mount, unmount, cleanup patterns
  • Event Listeners - React to session events
  • Session Management - Programmatic control
  • Quality Statistics - Monitor connection quality
  • Custom Themes - Theme customization
  • Virtual Backgrounds - Custom background images

📚 API Reference

Complete API documentation:

  • Core Methods (see skill.md)

    • joinSession() - Start a video session
    • closeSession() - End session and remove UI
    • destroy() - Clean up UI Toolkit instance
    • leaveSession() - Leave without destroying UI
  • Component Methods (see skill.md)

    • showControlsComponent() - Display control bar
    • showChatComponent() - Display chat panel
    • showUsersComponent() - Display participants list
    • showSettingsComponent() - Display settings panel
    • hideAllComponents() - Hide all components
  • Event Listeners (see skill.md)

    • onSessionJoined() - Session joined successfully
    • onSessionClosed() - Session ended
    • onSessionDestroyed() - UI Toolkit destroyed
    • onViewTypeChange() - View mode changed
    • on() - Subscribe to Video SDK events
    • off() - Unsubscribe from events
  • Information Methods (see skill.md)

    • getSessionInfo() - Get session details
    • getCurrentUserInfo() - Get current user
    • getAllUser() - Get all participants
    • getClient() - Get underlying Video SDK client
    • version() - Get version info
  • Control Methods (see skill.md)

    • changeViewType() - Switch view mode
    • mirrorVideo() - Mirror self video
    • isSupportCustomLayout() - Check device support
  • Statistics Methods (see skill.md)

    • subscribeAudioStatisticData() - Audio quality stats
    • subscribeVideoStatisticData() - Video quality stats
    • subscribeShareStatisticData() - Share quality stats

🔧 Configuration

  • Feature Configuration (see skill.md)

    • featuresOptions structure
    • Audio/Video options
    • Chat, Users, Settings
    • Virtual Background
    • Recording, Captions (paid features)
    • Theme customization
    • View modes
  • Session Configuration (see skill.md)

    • Required: videoSDKJWT, sessionName, userName
    • Optional: sessionPasscode, sessionIdleTimeoutMins
    • Debug mode
    • Web endpoint
    • Language settings

⚠️ Troubleshooting

Common Issues

  • React 18 peer dependency error
  • JWT token invalid
  • CSS not loading
  • Components not showing
  • Session join failures

See: troubleshooting/common-issues.md

Framework-Specific Issues

  • React: SSR, hydration, cleanup
  • Vue: Reactivity, lifecycle
  • Angular: Module imports, AOT
  • Next.js: App Router, basePath

Session Issues

  • Authentication failures
  • Connection problems
  • Video/audio not working
  • Screen share issues

📦 Sample Applications

Official Repositories:

FrameworkRepositoryKey Features
Reactvideosdk-ui-toolkit-react-sampleHooks, TypeScript
Vue.jsvideosdk-ui-toolkit-vuejs-sampleComposition API
Angularvideosdk-ui-toolkit-angular-sampleServices, Guards
JavaScriptvideosdk-ui-toolkit-javascript-sampleVanilla JS
Auth Endpointvideosdk-auth-endpoint-sampleNode.js JWT

🌐 External Resources

🎓 Learning Path

Beginner

  1. Read SKILL.md overview
  2. Follow Quick Start - Composite
  3. Generate JWT on server
  4. Join your first session
  5. Explore available features

Intermediate

  1. Try Component Mode
  2. Add event listeners
  3. Customize theme
  4. Add virtual backgrounds
  5. Integrate with your framework

Advanced

  1. Access underlying Video SDK
  2. Subscribe to quality statistics
  3. Handle all edge cases
  4. Implement custom layouts
  5. Build production-ready app

📋 Quick Reference Card

Minimal Working Example

javascript
import uitoolkit from "@zoom/videosdk-ui-toolkit";
import "@zoom/videosdk-ui-toolkit/dist/videosdk-ui-toolkit.css";

const config = {
  videoSDKJWT: "YOUR_JWT",
  sessionName: "test-session",
  userName: "User",
  featuresOptions: {
    video: { enable: true },
    audio: { enable: true }
  }
};

uitoolkit.joinSession(document.getElementById("container"), config);
uitoolkit.onSessionJoined(() => console.log("Joined"));
uitoolkit.onSessionClosed(() => uitoolkit.destroy());

Must-Remember Rules

  1. Always generate JWT server-side
  2. Always call destroy() on cleanup
  3. Always use React 18 (not 17/19)
  4. Always import CSS file
  5. Never expose SDK secret client-side
  6. Never skip onSessionClosed cleanup
  7. Never call components before joinSession

📞 Support


Navigation: ← Back to SKILL.md

Environment Variables

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Ui Toolkit Web AI skill do?

Zoom Video SDK UI Toolkit for Web. Use when building prebuilt or low-code Zoom Video SDK browser UI with @zoom/videosdk-ui-toolkit, composite UI, individual components, session lifecycle, React/Vue/Angular/Next.js/vanilla integration, featuresOptions, JWT auth, CSS/CDN install, troubleshooting, or deciding between UI Toolkit and raw Video SDK.

Why use Ui Toolkit Web on TypingMind?

Because you install it once and use it with any model. Ui Toolkit Web is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Ui Toolkit Web in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/zoom/skills/tree/main/skills/ui-toolkit. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Ui Toolkit Web?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Ui Toolkit Web?

As many as you like. As long as a model supports skills, you can use Ui Toolkit Web with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Ui Toolkit Web AI skill free?

Yes. It is published on GitHub by zoom under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇