Zoom Cobrowse Sdk logo

Zoom Cobrowse Sdk

Organization
zoom
zoom-cobrowse-sdk

Zoom Cobrowse SDK for web - JavaScript SDK for real-time collaborative browsing between agents and customers. Features include annotation tools, privacy masking, remote assist, and PIN-based session sharing.

Overview

Publisherzoom
Repositoryskills
Skill namezoom-cobrowse-sdk
Stars
78
Forks
16
Bundled files
32
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.

  • 32 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 Zoom Cobrowse Sdk 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/cobrowse-sdk .claude/skills/zoom-cobrowse-sdk
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Zoom Cobrowse Sdk 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 Zoom Cobrowse Sdk 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 Zoom Cobrowse Sdk 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 Cobrowse SDK - Web Development

Expert guidance for implementing collaborative browsing with the Zoom Cobrowse SDK. This SDK enables support agents to view and interact with a customer's browser session in real-time, with privacy controls and annotation tools.

Current public changelog release: 3.6.0 (June 29, 2026; verified 2026-07-10). The current 3.x line adds multi-agent remote-assist control, keyboard-accessible annotations, broader masking for image/SVG/iframe/canvas/text/background content, customer-only multi-tab transfer, annotation snap-to-element, and improved synchronization/latency. Pin and test an exact version before using a floating minor selector.

Official Documentation: https://developers.zoom.us/docs/cobrowse-sdk/
API Reference: https://marketplacefront.zoom.us/sdk/cobrowse/
Quickstart Repository: https://github.com/zoom/CobrowseSDK-Quickstart
Auth Endpoint Sample: https://github.com/zoom/cobrowsesdk-auth-endpoint-sample

Quick Links

New to Cobrowse SDK? Follow this path:

  1. Get Started Guide - Complete setup from credentials to first session
  2. Session Lifecycle - Understanding customer and agent flows
  3. JWT Authentication - Token generation and security
  4. Customer Integration - Integrate SDK into your website
  5. Agent Integration - Set up agent portal (iframe or npm)

Core Concepts:

Features:

Troubleshooting:

Reference:

SDK Overview

The Zoom Cobrowse SDK is a JavaScript library that provides:

  • Real-Time Co-Browsing: Agent sees customer's browser activity live
  • PIN-Based Sessions: Secure 6-digit PIN for customer-to-agent connection
  • Annotation Tools: Drawing, highlighting, vanishing pen, rectangle, color picker
  • Privacy Masking: CSS selector-based masking of sensitive form fields
  • Remote Assist: Agent can scroll customer's page (with consent)
  • Multi-Tab Persistence: Session continues when customer opens new tabs
  • Auto-Reconnection: Session recovers from page refresh (2-minute window)
  • Session Events: Real-time events for session state changes
  • HTTPS Required: Secure connections (HTTP only works on loopback/local development hosts)
  • No Plugins: Pure JavaScript, no browser extensions needed

Two Roles Architecture

Cobrowse has two distinct roles, each with different integration patterns:

Rolerole_typeIntegrationJWT RequiredPurpose
Customer1Website integration (CDN or npm)YesUser who shares their browser session
Agent2Iframe (CDN) or npm (BYOP only)YesSupport staff who views/assists customer

Key Insight: Customer and agent use different integration methods but the same JWT authentication pattern.

Read This First (Critical)

For customer/agent demos, treat the PIN from customer SDK event pincode_updated as the only user-facing PIN.

  • Show one clearly labeled value in UI (for example, Support PIN).
  • Use that same PIN for agent join.
  • Do not expose provisional/debug PINs from backend pre-start records to users.

If these rules are ignored, agent desk often fails with Pincode is not found / code 30308.

Typical Production Flow (Most Common)

This is the flow most teams implement first, and what users usually expect in demos:

  1. Customer starts session first (role_type=1)
    • Backend creates/records session
    • Backend returns customer JWT
    • Customer SDK starts and receives a PIN
  2. Agent joins second (role_type=2)
    • Agent enters customer PIN
    • Backend validates PIN and session state
    • Backend returns agent JWT
    • Agent opens Zoom-hosted desk iframe (or custom npm agent UI in BYOP)

If a demo only has one generic "session" user, it is incomplete for real cobrowse operations.

Prerequisites

Platform Requirements

  • Supported Browsers:

    • Chrome 80+ ✓
    • Firefox 78+ ✓
    • Safari 14+ ✓
    • Edge 80+ ✓
    • Internet Explorer ✗ (not supported)
  • Network Requirements:

    • HTTPS required (HTTP works on loopback/local development hosts only)
    • Allow cross-origin requests to *.zoom.us
    • CSP headers must allow Zoom domains (see CORS and CSP guide)
  • Third-Party Cookies:

    • Must enable third-party cookies for refresh reconnection
    • Privacy mode may limit certain features

Zoom Account Requirements

  1. Zoom Workplace Account with SDK Universal Credit
  2. Video SDK App created in Zoom Marketplace
  3. Cobrowse SDK Credentials from the app's Cobrowse tab

Note: Cobrowse SDK is a feature of Video SDK (not a separate product).

Need to create the Video SDK app first? Use Marketplace app management for app creation, manifest validation, Video SDK credential shape, and app-type quirks before generating Cobrowse JWTs.

Credentials Overview

You'll receive 4 credentials from Zoom Marketplace → Video SDK App → Cobrowse tab:

CredentialTypeUsed ForExposure Safe?
SDK KeyPublicCDN URL, JWT app_key claim✓ Yes (client-side)
SDK SecretPrivateSign JWTs✗ No (server-side only)
API KeyPrivateREST API calls (optional)✗ No (server-side only)
API SecretPrivateREST API calls (optional)✗ No (server-side only)

Critical: SDK Key is public (embedded in CDN URL), but SDK Secret must never be exposed client-side.

Quick Start

Step 1: Get SDK Credentials

  1. Go to Zoom Marketplace
  2. Open your Video SDK App (or create one)
  3. Navigate to the Cobrowse tab
  4. Copy your credentials:
    • SDK Key
    • SDK Secret
    • API Key (optional)
    • API Secret (optional)

Step 2: Set Up Token Server

Deploy a server-side endpoint to generate JWTs. Use the official sample:

bash
git clone https://github.com/zoom/cobrowsesdk-auth-endpoint-sample.git
cd cobrowsesdk-auth-endpoint-sample
npm install

# Create .env file
cat > .env << EOF
ZOOM_SDK_KEY=your_sdk_key_here
ZOOM_SDK_SECRET=your_sdk_secret_here
PORT=4000
EOF

npm start

Token endpoint:

javascript
// POST https://YOUR_TOKEN_SERVICE_BASE_URL
{
  "role": 1,           // 1 = customer, 2 = agent
  "userId": "user123",
  "userName": "John Doe"
}

// Response
{
  "token": "eyJhbGciOiJIUzI1NiIs..."
}

Step 3: Customer Side Integration (CDN)

html
<!DOCTYPE html>
<html>
<head>
  <title>Customer - Cobrowse Demo</title>
  <script type="module">
    const ZOOM_SDK_KEY = 'YOUR_SDK_KEY';
    
    // Load SDK from CDN
    (function(r, a, b, f, c, d) {
      r[f] = r[f] || { init: function() { r.ZoomCobrowseSDKInitArgs = arguments }};
      var fragment = a.createDocumentFragment();
      function loadJs(url) {
        c = a.createElement(b);
        d = a.getElementsByTagName(b)[0];
        c["async"] = false;
        c.src = url;
        fragment.appendChild(c);
      }
      loadJs(`https://us01-zcb.zoom.us/static/resource/sdk/${ZOOM_SDK_KEY}/js/3.6.0`);
      d.parentNode.insertBefore(fragment, d);
    })(window, document, "script", "ZoomCobrowseSDK");
  </script>
</head>
<body>
  <h1>Customer Support</h1>
  <button id="cobrowse-btn" disabled>Loading...</button>
  
  <!-- Sensitive fields - will be masked from agent -->
  <label>SSN: <input type="text" class="pii-mask" placeholder="XXX-XX-XXXX"></label>
  <label>Credit Card: <input type="text" class="pii-mask" placeholder="XXXX-XXXX-XXXX-XXXX"></label>
  
  <script type="module">
    let sessionRef = null;
    
    const settings = {
      allowAgentAnnotation: true,
      allowCustomerAnnotation: true,
      piiMask: {
        maskCssSelectors: ".pii-mask",
        maskType: "custom_input"
      }
    };
    
    ZoomCobrowseSDK.init(settings, function({ success, session, error }) {
      if (success) {
        sessionRef = session;
        
        // Listen for PIN code
        session.on("pincode_updated", (payload) => {
          console.log("PIN Code:", payload.pincode);
          // IMPORTANT: this is the PIN agent should use
          alert(`Share this PIN with agent: ${payload.pincode}`);
        });
        
        // Listen for session events
        session.on("session_started", () => console.log("Session started"));
        session.on("agent_joined", () => console.log("Agent joined"));
        session.on("agent_left", () => console.log("Agent left"));
        session.on("session_ended", () => console.log("Session ended"));
        
        document.getElementById("cobrowse-btn").disabled = false;
        document.getElementById("cobrowse-btn").innerText = "Start Cobrowse Session";
      } else {
        console.error("SDK init failed:", error);
      }
    });
    
    document.getElementById("cobrowse-btn").addEventListener("click", async () => {
      // Fetch JWT from your server
      const response = await fetch("https://YOUR_TOKEN_SERVICE_BASE_URL", {
        method: "POST",
        headers: { "Content-Type": "application/json" },
        body: JSON.stringify({
          role: 1,
          userId: "customer_" + Date.now(),
          userName: "Customer"
        })
      });
      const { token } = await response.json();
      
      // Start cobrowse session
      sessionRef.start({ sdkToken: token });
    });
  </script>
</body>
</html>

Step 4: Agent Side Integration (Iframe)

html
<!DOCTYPE html>
<html>
<head>
  <title>Agent Portal</title>
</head>
<body>
  <h1>Agent Portal</h1>
  <iframe 
    id="agent-iframe"
    width="1024" 
    height="768"
    allow="autoplay *; camera *; microphone *; display-capture *; geolocation *;"
  ></iframe>
  
  <script>
    async function connectAgent() {
      // Fetch JWT from your server
      const response = await fetch("https://YOUR_TOKEN_SERVICE_BASE_URL", {
        method: "POST",
        headers: { "Content-Type": "application/json" },
        body: JSON.stringify({
          role: 2,
          userId: "agent_" + Date.now(),
          userName: "Support Agent"
        })
      });
      const { token } = await response.json();
      
      // Load Zoom agent portal
      const iframe = document.getElementById("agent-iframe");
      iframe.src = `https://us01-zcb.zoom.us/sdkapi/zcb/frame-templates/desk?access_token=${token}`;
    }
    
    connectAgent();
  </script>
</body>
</html>

Step 5: Test the Integration

  1. Open two separate browsers (or incognito + normal)
  2. Customer browser: Open customer page, click "Start Cobrowse Session"
  3. Customer browser: Note the 6-digit PIN displayed
  4. Agent browser: Open agent page, enter the PIN code
  5. Both browsers: Session connects, agent can see customer's page
  6. Test features: Annotations, data masking, remote assist

Key Features

1. Annotation Tools

Both customer and agent can draw on the shared screen:

javascript
const settings = {
  allowAgentAnnotation: true,      // Agent can draw
  allowCustomerAnnotation: true    // Customer can draw
};

Available tools:

  • Pen (persistent)
  • Vanishing pen (disappears after 4 seconds)
  • Rectangle
  • Color picker
  • Eraser
  • Undo/Redo

2. Privacy Masking

Hide sensitive fields from agents using CSS selectors:

javascript
const settings = {
  piiMask: {
    maskType: "custom_input",           // Mask specific fields
    maskCssSelectors: ".pii-mask, #ssn", // CSS selectors
    maskHTMLAttributes: "data-sensitive=true" // HTML attributes
  }
};

Supported masking:

  • Text nodes ✓
  • Form inputs ✓
  • Select elements ✓
  • Images ✗ (not supported)
  • Links ✗ (not supported)

3. Remote Assist

Agent can scroll the customer's page:

javascript
const settings = {
  remoteAssist: {
    enable: true,
    enableCustomerConsent: true,        // Customer must approve
    remoteAssistTypes: ['scroll_page'], // Only scroll supported
    requireStopConfirmation: false      // Confirmation when stopping
  }
};

4. Multi-Tab Session Persistence

Session continues when customer opens new tabs:

javascript
const settings = {
  multiTabSessionPersistence: {
    enable: true,
    stateCookieKey: '$$ZCB_SESSION$$'  // Cookie key (base64 encoded)
  }
};

Session Lifecycle

Customer Flow

  1. Load SDK → CDN script loads ZoomCobrowseSDK
  2. InitializeZoomCobrowseSDK.init(settings, callback)
  3. Fetch JWT → Request token from your server (role_type=1)
  4. Start Sessionsession.start({ sdkToken })
  5. PIN Generatedpincode_updated event fires
  6. Share PIN → Customer gives 6-digit PIN to agent
  7. Agent Joinsagent_joined event fires
  8. Session Active → Real-time synchronization begins
  9. End Sessionsession.end() or agent leaves

Agent Flow

  1. Fetch JWT → Request token from your server (role_type=2)
  2. Load Iframe → Point to Zoom agent portal with token
  3. Enter PIN → Agent inputs customer's 6-digit PIN
  4. Connectsession_joined event fires
  5. View Session → Agent sees customer's browser
  6. Use Tools → Annotations, remote assist, zoom
  7. Leave Session → Click "Leave Cobrowse" button

Session Recovery (Auto-Reconnect)

When customer refreshes the page:

javascript
ZoomCobrowseSDK.init(settings, function({ success, session, error }) {
  if (success) {
    const sessionInfo = session.getSessionInfo();
    
    // Check if session is recoverable
    if (sessionInfo.sessionStatus === 'session_recoverable') {
      session.join();  // Auto-rejoin previous session
    } else {
      // Start new session
      session.start({ sdkToken });
    }
  }
});

Recovery window: 2 minutes. After 2 minutes, session ends.

Critical Gotchas and Best Practices

⚠️ CRITICAL: SDK Secret Must Stay Server-Side

Problem: Developers often accidentally embed SDK Secret in frontend code.

Solution:

  • SDK Key → Safe to expose (embedded in CDN URL)
  • SDK Secret → Never expose (use for JWT signing server-side)
javascript
// ❌ WRONG - Secret exposed in frontend
const jwt = signJWT(payload, 'YOUR_SDK_SECRET');  // Security risk!

// ✅ CORRECT - Secret stays on server
const response = await fetch('/api/token', {
  method: 'POST',
  body: JSON.stringify({ role: 1, userId, userName })
});
const { token } = await response.json();

SDK Key vs API Key (Different Purposes!)

CredentialUsed ForJWT Claim
SDK KeyCDN URL, JWT app_keyapp_key: "SDK_KEY"
API KeyREST API calls (optional)Not used in JWT

Common mistake: Using API Key instead of SDK Key in JWT app_key claim.

Session Limits

LimitValueWhat Happens
Customers per session1Error 1012: SESSION_CUSTOMER_COUNT_LIMIT
Agents per session5Error 1013: SESSION_AGENT_COUNT_LIMIT
Active sessions per browser1Error 1004: SESSION_COUNT_LIMIT
PIN code length10 chars maxError 1008: SESSION_PIN_INVALID_FORMAT

Session Timeout Behavior

EventTimeoutWhat Happens
Agent waiting for customer3 minutesSession ends automatically
Page refresh reconnection2 minutesSession ends if not reconnected
Reconnection attempts2 times maxSession ends after 2 failed attempts

HTTPS Requirement

Problem: SDK doesn't load on HTTP sites.

Solution:

  • Production: Use HTTPS ✓
  • Development: Use a loopback host for local HTTP testing ✓
  • Development: Use a local HTTPS endpoint with a trusted/self-signed cert if required ✓

Third-Party Cookies Required

Problem: Refresh reconnection doesn't work.

Solution: Enable third-party cookies in browser settings.

Affected scenarios:

  • Browser privacy mode
  • Safari with "Prevent cross-site tracking" enabled
  • Chrome with "Block third-party cookies" enabled

Distribution Method Confusion

MethodUse CaseAgent IntegrationBYOP Required
CDNMost use casesZoom-hosted iframeNo (auto PIN)
npmCustom agent UI, full controlCustom npm integrationYes (required)

Key Insight: If you want npm integration, you must use BYOP (Bring Your Own PIN) mode.

Cross-Origin Iframe Handling

Problem: Cobrowse doesn't work in cross-origin iframes.

Solution: Inject SDK snippet into cross-origin iframes:

html
<script>
const ZOOM_SDK_KEY = "YOUR_SDK_KEY_HERE";

(function(r,a,b,f,c,d){r[f]=r[f]||{init:function(){r.ZoomCobrowseSDKInitArgs=arguments}};
var fragment=a.createDocumentFragment();function loadJs(url) {c=a.createElement(b);d=a.getElementsByTagName(b)[0];c.async=false;c.src=url;fragment.appendChild(c);};
loadJs('https://us01-zcb.zoom.us/static/resource/sdk/${ZOOM_SDK_KEY}/js');d.parentNode.insertBefore(fragment,d);})(window,document,'script','ZoomCobrowseSDK');
</script>

Same-origin iframes: No extra setup needed.

Known Limitations

Synchronization Limits

Not synchronized:

  • HTML5 Canvas elements
  • WebGL content
  • Audio and Video elements
  • Shadow DOM
  • PDF rendered with Canvas
  • Web Components

Partially synchronized:

  • Drop-down boxes (only selected result)
  • Date pickers (only selected result)
  • Color pickers (only selected result)

Rendering Limits

  • High-resolution images may be compressed
  • Different screen sizes may cause CSS media query differences
  • Cross-origin images may not render (CORS restrictions)
  • Cross-origin fonts may not render (CORS restrictions)

Masking Limits

Supported:

  • Text nodes ✓
  • Form inputs ✓
  • Select elements ✓

Not supported:

  • <img> elements ✗
  • Links ✗

Complete Documentation Library

This skill includes comprehensive guides organized by category:

Core Concepts

Examples

References

Troubleshooting

Resources


Need help? Start with Integrated Index section below for complete navigation.


Integrated Index

This section was migrated from SKILL.md.

Complete navigation guide for all Cobrowse SDK documentation.

Getting Started (Start Here!)

If you're new to Zoom Cobrowse SDK, follow this learning path:

  1. SKILL.md - Main overview and quick start
  2. 5-Minute Runbook - Preflight checks for common failures
  3. Get Started Guide - Step-by-step setup from credentials to first session
  4. Session Lifecycle - Understand the complete customer and agent flow
  5. Customer Integration - Integrate SDK into your website
  6. Agent Integration - Set up agent portal

Core Concepts

Foundational concepts you need to understand:

Examples and Patterns

Complete working examples for common scenarios:

Session Management

Features

References

Complete API and configuration references:

SDK Reference

  • API Reference - All SDK methods and interfaces

    • ZoomCobrowseSDK.init()
    • session.start()
    • session.join()
    • session.end()
    • session.on()
    • session.getSessionInfo()
  • Settings Reference - All initialization settings

    • allowAgentAnnotation
    • allowCustomerAnnotation
    • piiMask
    • remoteAssist
    • multiTabSessionPersistence
  • Session Events Reference - All event types

    • pincode_updated
    • session_started
    • session_ended
    • agent_joined
    • agent_left
    • session_error
    • session_reconnecting
    • remote_assist_started
    • remote_assist_stopped

Error Reference

  • Error Codes - Complete error code reference
    • 1001-1017: Session errors
    • 2001: Token errors
    • 9999: Service errors

Official Documentation

Troubleshooting

Quick diagnostics and common issue resolution:

  • Common Issues - Quick fixes for frequent problems

    • SDK not loading
    • Token generation fails
    • Agent can't connect
    • Fields not masked
    • Session doesn't reconnect after refresh
  • Error Codes - Error code lookup and solutions

    • Session start/join failures (1001, 1011, 1016)
    • Session limit errors (1002, 1004, 1012, 1013, 1015)
    • PIN code errors (1006, 1008, 1009, 1010)
    • Token errors (2001)
  • CORS and CSP - Cross-origin and Content Security Policy setup

    • Access-Control-Allow-Origin headers
    • Content-Security-Policy headers
    • Cross-origin iframe handling
    • Same-origin iframe handling
  • Browser Compatibility - Browser requirements and limitations

    • Supported browsers (Chrome 80+, Firefox 78+, Safari 14+, Edge 80+)
    • Internet Explorer not supported
    • Privacy mode limitations
    • Third-party cookie requirements

By Use Case

Find documentation by what you're trying to do:

I want to...

Set up cobrowse for the first time:

Add annotation tools:

Hide sensitive data from agents:

Let agents control customer's page:

Use custom PIN codes:

Handle page refreshes:

Integrate with npm (not CDN):

Debug session connection issues:

Configure CORS and CSP headers:

By Error Code

Quick lookup for error code solutions:

Session Errors

PIN Errors

Auth Errors

Service Errors

Official Resources

External documentation and samples:

Documentation Structure

cobrowse-sdk/
├── SKILL.md                    # Main skill entry point
├── SKILL.md                    # This file - complete navigation
├── get-started.md              # Step-by-step setup guide
├── concepts/                   # Core concepts
│   ├── two-roles-pattern.md
│   ├── session-lifecycle.md
│   ├── jwt-authentication.md
│   └── distribution-methods.md
├── examples/                   # Working examples
│   ├── customer-integration.md
│   ├── agent-integration.md
│   ├── annotations.md
│   ├── privacy-masking.md
│   ├── remote-assist.md
│   ├── multi-tab-persistence.md
│   ├── byop-custom-pin.md
│   ├── session-events.md
│   └── auto-reconnection.md
├── references/                 # API and config references
│   ├── api-reference.md        # SDK methods
│   ├── settings-reference.md   # Init settings
│   ├── session-events.md       # Event types
│   ├── error-codes.md          # Error reference
│   ├── get-started.md          # Official docs (crawled)
│   ├── features.md             # Official docs (crawled)
│   ├── authorization.md        # Official docs (crawled)
│   └── api.md                  # API docs (crawled)
└── troubleshooting/            # Problem resolution
    ├── common-issues.md
    ├── error-codes.md
    ├── cors-csp.md
    └── browser-compatibility.md

Search Tips

Find by keyword:


Not finding what you need? Check the Official Documentation or ask on the Dev Forum.

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 Zoom Cobrowse Sdk AI skill do?

Zoom Cobrowse SDK for web - JavaScript SDK for real-time collaborative browsing between agents and customers. Features include annotation tools, privacy masking, remote assist, and PIN-based session sharing.

Why use Zoom Cobrowse Sdk on TypingMind?

Because you install it once and use it with any model. Zoom Cobrowse Sdk 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 Zoom Cobrowse Sdk in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/zoom/skills/tree/main/skills/cobrowse-sdk. 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 Zoom Cobrowse Sdk?

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 Zoom Cobrowse Sdk?

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

Is the Zoom Cobrowse Sdk 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 👇