Conversational AI
  • Introduction
  • Getting started
    • Getting started
    • Adding content to your bot
    • Capture information with entities
    • Capture information with input validation
    • Reusing intents with context
    • Flow navigation with variables
    • Adding new users to your account
  • Understanding users
    • Natural Language Processing (NLP)
      • NLP threshold
      • NLP Import & Export
      • Train your bot with actual user messages
      • NLP Dashboard & NLP Improve
      • Synonym entities
      • System entities
      • Supported languages
      • Intent templates
    • Expression generator
    • Context
    • Multi-language bots
  • Bot answers
    • Bot dialogs
      • Message components
      • Go To
      • Input Validation
      • Action
      • Translations
    • Conversations
    • Analytics
      • User flow
    • Publishing your bot
    • Events
    • Reuse flows
    • Settings
      • Variables
  • Integrations
    • API integration
      • Advanced API integrations
    • Chat message structure for API's
    • Retrieving data from Airtable (GET)
    • Sending data to Airtable (POST)
    • Human handover & live chat
      • #Interact
      • RingCentral Engage Digital
      • Genesys Cloud
      • Help Scout
      • Zendesk Chat
      • Intercom
      • Sparkcentral (beta)
      • Offloading Webhook
  • Channels
    • Channels
    • Facebook Messenger
      • Facebook Admin Removal
      • Facebook Webview Whitelisting
    • WhatsApp Business API
    • Google Assistant
    • Webhook Channel API
    • Chat widget
    • Phone & voice
    • Workplace from Facebook
    • Sinch Conversation API (beta)
  • Tips & Best practices
    • How to NLP
    • Creating diverse expressions
    • Why is my bot not responding the way I want it to?
    • What makes a good chatbot?
    • How to recognize a returning bot user
    • Gathering user feedback
    • Using time in your chatbot
Powered by GitBook
On this page
  • Creating your chat widget
  • White listing
  • Embedding the web widget on your website
  • Should I use the SDK or an iframe?
  • SDK
  • Iframe
  • Encryption
  • Live example

Was this helpful?

  1. Channels

Chat widget

PreviousWebhook Channel APINextPhone & voice

Last updated 4 years ago

Was this helpful?

One of the most common ways of making a bot available to your clients is through a chat widget. In this technical guide, we will guide you through the process of integrating the bot in your website.

Creating your chat widget

To create a chat widget for your bot, navigate to the Channels tab and then click on the pencil icon next to Web.

Click on "+ Create" to start building your first custom chat widget.

You can create multiple chat widgets for one bot, remove them, view them, and copy them using the buttons in the chat widget table.

On the Chat Widget page you can customize a number of key chat widget components:

  • Config tab

    • Title of the bot

    • Template

    • Upload custom CSS

    • Title font & font size

    • Paragraph font & font size

    • White listing of the domains you want the bot to be activated on (i.e. your own website domain and any other domains you use for testing)

    • Images (title, bot, user, send icon, SDK button - which is the image that's shown when the bot is closed)

  • Colors tab

    • Colors of all the key components of the chatbot

  • Translations tab

    • Translations for the default chat widget messages such as "Write reply..."

White listing

To make sure your bot can only be activated on your website, all Chatlayer.ai chat widgets must be whitelisted before use. To do this, go to the config tab of the Chat Widget builder and add your domains in regex format.

Embedding the web widget on your website

To embed the bot into your website, start by clicking the Embed button in the top right of the Custom Chat Widget page.

There are two ways of integrating the Chatlayer.ai Chat Widget onto your website: you can use an iframe or the SDK.

You can switch between SDK and iframe using the Type switcher above your widget:

Should I use the SDK or an iframe?

The Chatlayer chat widget SDK is a layer on top of the iframe, which includes some other functionalities such as a button that is shown before the chat window is opened, an option to close the chatbot.

If you want to get the chat widget onto your website with minimal custom development it's best to use the SDK. The only thing you need to do is include a HTML script tag as described below.

If you want more control over other elements, such as the chat button, it's best to use the iframe.

SDK

You can load the Chatlayer.ai chat widget by using the script tag below. Calling the chatlayer function will render the button and your chat widget to the page.

<script src="https://chatbox.staging.chatlayer.ai/sdk/INSERT-TOKEN-HERE" 
        onload='chatlayer()' async>
    // The token can be found in the Chat Widget configuration window
</script>

Remove .staging from the URL if you want to integrate a production bot. You can add parameters to the chatlayer function to include additional functionality.

<script src="https://chatbox.staging.chatlayer.ai/sdk/INSERT-TOKEN-HERE" 
        onload='chatlayer({"lang":"en","animation":"grow"})' async>
</script>

In this case, the bot will be opened in English, and when a user clicks the SDK button, it will be opened with the Grow animation.

SDK Options

Property name

Type

Remark

withCloseButton

boolean

The close button allows customers to close the bot using an "X" icon in the top right of the chat widget

autoOpen

boolean

noButton

boolean

session

object

Add data to the session which can be used to guide the flow.

sessionId

string

A session ID can be used to continue the conversation after the user has left the page. Make sure it's 20 characters or longer.

language

string

button

HTMLElement

wrapper

HTMLElement

Passing a variable to the SDK

If you want to pass data from your website to the chatbot, you can add these to the chatlayer function. All data will be put onto the root.session variable in the Chatlayer.ai session. This data can then be used to customize the chatbot flow based on website data.

<script src="https://chatbox.staging.chatlayer.ai/sdk/INSERT-TOKEN-HERE"
 onload='chatlayer({session: {number: 1234, firstName: "User", lastName: "Name" }})' async>
</script>

Passing the variables firstName and lastName to the SDK allows you to set the corresponding internal Chatlayer session variables. These specific variables decide how the user will show up in the Chatlayer conversations table.

internal: {
  ...
  user: {
      firstName: 'User',
      lastName: 'Name',
        },
        ...
    },
    ...
 session: {
     number: 1234,
     firstName: 'User',
     lastName: 'Name'
    }
    ...
    firstname: 'User',
    lastname: 'Name'
}

User authentication with the chat widget SDK

For a lot of use cases, it's required to know who is talking to the bot. One way to do this is by sending a login ID, gathered on the website where your user is logged in, and pass that on to the bot, when the SDK is initialized:

<script src="YOUR_CHATWIDGET_URL" 
onload='chatlayer({ sessionId: LoginID }})' async>
</script>

Chat widget button behaviour

If you import the chat widget with our SDK, you can decide when the widget opens or closes. An example:

var chatlayerChat = chatlayer({ lang: 'en'});
// This creates a Chatlayer.ai instance and adds the icon to your webpage

chatlayerChat.open();
// Opens the Chatlayer.ai chat window
setTimeout(()=>chatlayerChat.open(),10000)
// Opens the Chatlayer.ai chat window with a predefined delay
chatlayerChat.toggle();
// Toggles between the opened and closed version of the chat widget
chatlayerChat.close();
// Minimizes the chat widget window to the button

Make sure that the SDK script is loaded before running these functions.

Iframe

Embedding the chat widget in an iframe is the most straight forward embedding option. You determine how and where you place the element and you can style it as you like.

<iframe src="https://chatbox.staging.chatlayer.ai/INSERT-TOKEN-HERE?language=en"/>

Remove .staging from the URL if you want to integrate a production bot. You can change ?language=en to any language the bot supports.

Defining the sender ID

If you want to recognize a returning bot user, you can send a unique sender ID for each person opening the bot, which can then be used to open the same conversation when the page is reloaded.

<iframe src="https://chatbox.staging.chatlayer.ai/TOKEN?senderID=5241bbd5-6fd8-4333-85e0-72bsdf04f4c91"/>

Make sure that your sender ID is 20 characters or longer.

Passing a variable to the iframe

The same goal can also be achieved by changing the iframe url:

<iframe src="https://chatbox.staging.chatlayer.ai/TOKEN?senderId=12345TEST&session[data][first_name]=Charlie"/>

The example above, would result in a session containing:

{
   "id":"1234TEST",
   "session":{
      "data":{
         "first_name":"Charlie"
      }
   }
}

Size and position of iframe

By adding the properties underneath to your CSS, you can resize the chat widget window. You can use these parameters to make the chat widget responsive for mobile devices.

<style>
    iframe.chatlayer-iframe {
        width: 350px
        max-width: 100%;
        max-height: 560px;
    }
</style>

Encryption

There are two main options for securing the chat widget. You can either verify the payload by using JWT, making sure it hasn't been tampered with, or you can encrypt the session created by the chat widget by using AES-256 data protection.

When you turn on the AES-256 data protection, you can only pass an encrypted token as the chat session. This token should be generated in your own back-end. The code to generate the token looks like this (using node.js):

const crypto = require('crypto');
const SECRET = 'CHATLAYER_TOKEN'
const createAESToken = (payload, secret = SECRET) => {
    const iv = Buffer.from(crypto.randomBytes(8)).toString('hex');
    const secretBuffer = Buffer.from(secret, 'hex');
    const cipher = crypto.createCipheriv('aes-256-cbc', secretBuffer, iv);
    const update = cipher.update(typeof payload === 'object' ? JSON.stringify(payload) : String(payload));
    const encrypted = Buffer.concat([update, cipher.final()]);
    return iv + encrypted.toString('hex');
};
const token = createAESToken({
    exp: Math.floor(Date.now() / 1000) + 36000,
    session: {
        isAuthorized: true
    },
    sessionId: 'UNIQUE_ID_FOR_THIS_USER'
});

How you pass this token to the client will depend on how you are generating your html, but this server-generated token should be what is passed to the chatlayer function:const chat = chatlayer({token: serverGeneratedToken});

Live example

We've created an example of how you can initialize and destroy the Chatlayer.ai chat widget through your own Javascript code.

<html>
  <head>
    <title>Chatlayer Events Example</title>
    <meta charset="UTF-8" />
  </head>

  <body>
    <button onclick="initChatlayer()">Initialize Chatlayer</button>
    <button onclick="destroyChatlayer()">Destroy Chatlayer</button>
    <br />
    <br />
    <button onclick="openChatlayer()">Open Chatlayer Widget</button>
    <button onclick="closeChatlayer()">Close Chatlayer Widget</button>

    <script>
      /**
       * This example showcases the destruction of a Chatlayer webwidget
       * by triggering the JSON builder plugin in the flow.
       *
       * The JSON builder action dialog was set up as follows:
       * https://minio.dev.chatlayer.ai/public/JSONBuilder.png
       *
       */

      let chatlayerInstance = null;

      const initChatlayer = () => {
        if (chatlayerInstance) return;
        console.log("Initializing the Chatlayer widget ...");
        chatlayerInstance = chatlayer({ lang: "en" });
        // Event types are "event", "open" and "close"

        chatlayerInstance.on("open", event => {
          console.log('OPENED');
        });

        chatlayerInstance.on("close", event => {
          console.log('CLOSED')
        });

        chatlayerInstance.on("event", event => {
          // Note the content of event here: the "action" key with value "destroy"
          // originates directly from the configured JSON Builder action.
          const isChatlayerDestroyEvent = event.action === "destroy";
          if (isChatlayerDestroyEvent) {
            destroyChatlayer();
          }
        });
      };

      const destroyChatlayer = () => {
        if (!chatlayerInstance) return;
        console.log("Destroying the Chatlayer widget ...");
        // To destroy the widget, it has to be opened first
        chatlayerInstance.open();
        chatlayerInstance.destroy();
        chatlayerInstance = null;
      };

      const openChatlayer = () => {
        if (!chatlayerInstance) {
          console.log("Please initialize Chatlayer first!");
          return;
        }
        chatlayerInstance.open();
      };

      const closeChatlayer = () => {
        if (!chatlayerInstance) {
          console.log("Please initialize Chatlayer first!");
          return;
        }
        chatlayerInstance.close();
      };
    </script>

    <!-- Note the absence of the onload prop on the following script tag. -->
    <!-- To automatically initialize the webwidget, you would add the prop: -->
    <!-- onload="initChatlayer()" -->
    <script
      src="https://chatbox.staging.chatlayer.ai/sdk/5ecbcf1514c3dc4942f05f96"
      async
    ></script>
  </body>
</html>

JSON builder plugin configuration:

As you can see from above example, you can initialize the widget, open/close it as well as destroy it from your own code. The example shows how you can use the JSON Builder plugin to trigger a destroy event for the chat widget.

An example of this regex format for : ^https:\/\/www\.chatlayer\.ai(\/|$)

This login ID can then be reused to perform calls with the .

A user's session can be updated at any time through the .

View a live version of this code .

https://www.chatlayer.ai
API plugin
here
Asynchronous Actions API