The vf-chatbot-modal
component appears as an overlay, typically in the bottom right corner of the screen. It is designed for contextual assistance, allowing users to receive support without navigating away from the current page.
Only use the chatbot modal if it truly enables it helps users:
For example, the modal works well during checkouts or while filling out complex forms, where users may need immediate support without losing their progress. Chatbot modals can also benefit users who are familiar with a service and want targeted help to speed up their work. Test with users to ensure the embedded assistance enhances rather than disrupts their workflow.
The chatbot modal overlays the existing interface and may distract users or obscure important content. Not all users will notice the chatbot icon or understand when to interact with it.
Do not use the chatbot modal if the help content is essential for completing a task. Critical guidance should be directly visible in the interface.
Test your workflow without a chatbot first. Clear content, in-line guidance, and intuitive layout may negate the need for a modal.
It's usually better to:
Avoid putting the chatbot modal for help or multi-step interactions. These are better suited to a chatbot standalone experience or a well-structured support section.
Avoid placing modals over elements that users need to interact with, and never trigger the chatbot automatically without user intent.
The chatbot modal and chatbot standalone are two distinct variants for delivering conversational interfaces. Choose between them based on use case needs and task complexity.
Element | Description |
---|---|
Title bar | Shows the chatbot name, minimise button and close button. The title bar may include a dropdown for selecting categories. |
Dialogue section | Scrollable chat window showing the conversation log. |
Intro message | A brief onboarding message explaining the purpose and capabilities of the chatbot. Shows the icon, title and short message (Max. 2 lines of text). |
Banner | Used to show optional disclaimers or alerts (Max. 3 lines of text). For cases that require user consent, use a blur overlay on the background or a pre-access popup instead, as banners may be missed. |
Text input area | Open input field for typing and sending queries. Expands up to 5 lines, after which it becomes scrollable. |
Flow | Details |
---|---|
Suggested prompts | Appear on the initial screen. Provide a quick-start to users with clickable queries. They help to provide context on the type of questions the user can ask on the platform. (Max. 60 characters) |
Closing the chat dialogue | Clicking on the close icon "X" triggers a confirmation prompt to prevent accidental loss of the chat log. |
Error management | If the chatbot is unable to provide a response to the query, display a clear error message and provide an alternative way for them to get their answers. |
Links | Displayed in a clear underlined style. They can be shown inline or as a list. |
Source attribution | Chips are shown in relevant paragraphs which cite the sources. Links to the sources and more details can be accessed via a "View sources" button. |
Feedback on a query level | Users can assess the AI responses with a thumbs up/thumbs down or optional close/open response fields to give more details. |
Category selection | A dropdown in the title bar lets users switch focus areas (e.g LLM version or data source). Single or multi-selection variants can be used depending on the use case. |
You can access working example of modal version by clicking on the floating action button at the bottom right of this page.
The Chatbot branding elements follow EMBL brand guidelines but can be updated to suit your use case. For advice on branding updates please contact the EMBL Communications Team.
Texts shown in the examples are placeholder content. Please review and update all wording to fit your your project needs and ensure it meets legal, accessibility and organisational requirements.
The component targets WCAG 2.1 AA accessibility standard.
This repository is distributed with npm. After installing npm and yarn, you can install vf-chatbot-modal
and other dependant components with this command.
yarn add @visual-framework/vf-chatbot @visual-framework/vf-chatbot-modal @visual-framework/vf-chatbot-fab @visual-framework/vf-chatbot-dialog @visual-framework/vf-chatbot-action-prompt @visual-framework/vf-chatbot-feedback @visual-framework/vf-chatbot-prompt @visual-framework/vf-chatbot-selector @visual-framework/vf-chatbot-sources @visual-framework/vf-chatbot-welcome
@import "@visual-framework/vf-chatbot-modal/index.scss";
import { initVFChatbot } from 'vf-chatbot/vf-chatbot.js';
import { initVFChatbotModal } from 'vf-chatbot-modal/vf-chatbot-modal.js';
import { initVFChatbotFab } from 'vf-chatbot-fab/vf-chatbot-fab.js';
window.addEventListener("load", () => {
initVFChatbot(config);
});
where config is the configuration object with different options as described below
const config = {
// Basic Settings
type: "modal",
chatbotBottomMargin: 100, //This is optional and is used to apply bottom margin in pixels
title: "AI Assistant",
welcome_message: "Welcome! I'm here to help",
input_placeholder: "Ask me anything...",
welcome_max_suggestions: 4,
// Content & Branding
disclaimer: 'Custom disclaimer with <a href="/privacy">privacy policy</a>',
footnote: 'Custom footnote with <a href="/feedback">feedback link</a>',
// Icons & Assets
icons: {
assistant_avatar: "path/to/assistant-icon.svg",
user_avatar: "path/to/user-icon.svg",
send_button: "path/to/send-icon.svg",
main_logo_url: "path/to/logo.svg",
selector_logo_url: "path/to/selector-logo.svg",
},
// API Configuration
api: {
chat_endpoint: "/api/chat",
feedback_endpoint: "/api/feedback",
qa_data_url: "path/to/qa-data.json",
headers: {
"Content-Type": "application/json",
"Authorization": "Bearer your-token"
},
timeout: 10000
},
// Feature Toggles
features: {
enable_welcome: true,
enable_feedback: true,
enable_sources: true,
enable_sources_custom_format: true,
enable_welcome_suggestions: true,
enable_typing_indicator: true,
enable_disclaimer: true,
enable_predefined_qa: true,
enable_fallback_responses: true,
enable_qa_data_loading: true,
enable_instant_feedback: false
},
// New options
behavior: {
auto_scroll: true,
typing_delay: 800,
show_scrollbar: false // or false to hide scrollbar
},
selectorContext: {
chatbotRoutes: {
multiSelect: false,
showSearch: true,
showSearchThreshold: 5,
routes: "assets/vf-chatbot-selector-services.json",
placeholder: "Search",
title: "Services"
}
},
handlers: {
on_message_send: "handleMessageSend",
on_response_receive: "handleResponseReceive",
on_feedback_submit: "handleFeedbackSubmit",
on_suggestion_click: "handleSuggestionClick",
on_error: "handleError",
on_conversation_start: "handleConversationStart",
on_conversation_end: "handleConversationEnd"
}
};
The chatbot includes an integrated selector component which can be configured to present different selection options to user:
const selectorConfig = {
selectorContext: {
chatbotRoutes: {
// Multi-selection settings
multiSelect: true,
maxMultiSelect: 3,
// Search functionality
showSearch: true,
showSearchThreshold: 5,
// "All Services" option
showAllServices: true,
showAllServicesSelected: true,
// Data source
routes: "assets/vf-chatbot-selector-services.json",
// UI labels
placeholder: "Search",
title: "Available Services"
}
}
};
Selector Data Format (JSON):
{
"routes": [
{
"id": "service-1",
"title": "Service 1 title"
},
{
"id": "service-2",
"title": "Service 2 title"
}
]
}
Chatbot comes with a provision to allow custom event handlers. These handlers can be defined in your code to handle specific actions for different events triggered during interaction with chatbot. Configure custom handlers for chatbot events:
const config = {
handlers: {
on_message_send: "handleMessageSend",
on_response_receive: "handleResponseReceive",
on_feedback_submit: "handleFeedbackSubmit",
on_suggestion_click: "handleSuggestionClick",
on_error: "handleError",
on_conversation_start: "handleConversationStart",
on_conversation_end: "handleConversationEnd"
}
};
// Implement handler functions
function handleMessageSend(message, conversationId) {
console.log('User sent:', message);
// Track analytics, log conversations, etc.
}
function handleResponseReceive(response, sources, prompts) {
console.log('Assistant responded:', response);
// Process response, update UI, etc.
}
function handleFeedbackSubmit(feedbackData) {
console.log('Feedback received:', feedbackData);
// Send to analytics, update models, etc.
}
Likewise you can also listen to events emitted by the chatbot for specific interactions:
// Listen for specific chatbot events
document.addEventListener('vf-chatbot:message-send', (event) => {
const { message, conversationId } = event.detail;
// Handle message send
});
document.addEventListener('vf-chatbot:message-receive', (event) => {
const { message, conversationId } = event.detail;
// Handle message send
});
document.addEventListener('vf-chatbot-feedback:submit', (event) => {
const { messageId, feedbackType, feedbackText, feedbackComment } = event.detail;
// Handle feedback submission
});
document.addEventListener('vf-chatbot-welcome:suggestion-click', (event) => {
const { question } = event.detail;
// Handle suggestion clicks
});
document.addEventListener('vf-chatbot:assistant-change', (obj) => {
const { selectedRoutes } = obj.selectedAssistants;
const { conversationId } = obj.conversationId;
// Handle service selection
});
document.addEventListener('vf-chatbot:error', (event) => {
const { message, conversationId } = event.detail;
// Handle message send
});
document.addEventListener('vf-chatbot:conversation-start', (event) => {
const { message, conversationId } = event.detail;
// Handle conversation start
});
document.addEventListener('vf-chatbot:conversation-end', (event) => {
const { message, conversationId } = event.detail;
// Handle conversation end
});
const config = {
features: {
enable_feedback: true,
enable_instant_feedback: false // Default
}
};
const config = {
features: {
enable_feedback: true,
enable_instant_feedback: true // Thumbs up/down only
}
};
Load predefined questions and answers:
const config = {
features: {
enable_predefined_qa: true,
enable_qa_data_loading: true
},
api: {
qa_data_url: "path/to/qa-data.json"
}
};
Q&A Data Format:
{
"predefinedQA": {
"How can I submit genomic data to EMBL-EBI?": {
"answer": "To submit genomic data, visit the EMBL-EBI submission portal, where you’ll find step-by-step guides and tools for submitting sequencing data, assemblies, annotations, and more.",
"sources": [
{
"domain": "ebi.ac.uk",
"title": "EMBL's European Bioinformatics Institute",
"url": "https://www.ebi.ac.uk/",
"description": "Run BLAST searches against comprehensive sequence databases at EMBL-EBI."
},
{
"domain": "ena-docs.readthedocs.io",
"title": "ENA Documentation",
"url": "https://ena-docs.readthedocs.io/en/latest/",
"description": "ENA Documentation"
}
]
}
},
"fallbackResponses": [
{
"answer": "I'm sorry, I'm having trouble connecting to my knowledge base right now. Could you try again in a moment?",
"prompts": [
{
"action_text": "Contact support team",
"action_url": "tel:+44 1223 494 444"
},
{
"action_text": "Submit a support request",
"action_url": "https://www.ebi.ac.uk/about/contact/support/"
}
]
}
]
}
Your chat API should accept POST requests:
// Request format
{
"message": "User's question",
"conversationId": "unique-id",
"context": {
"selectedServices": ["service-1", "service-2"]
}
}
// Response format
{
"response": "Assistant's answer",
"sources": [
{
"title": "Documentation Link",
"url": "https://example.com/docs"
}
],
"prompts": [
{
"action_text": "Learn More",
"action_url": "https://example.com/learn"
}
]
}
const config = {
features: {
enable_welcome: true,
enable_welcome_suggestions: true
},
welcome_logo: true,
welcome_message: "Welcome to our AI assistant!",
welcome_suggestions_title: "Popular questions:",
welcome_max_suggestions: 6
};
const config = {
features: {
enable_sources: true
enable_sources_custom_format: true,
}
};
// Sources in API response
{
"response": "Here's the information...",
"sources": [
{
"title": "Official Documentation",
"url": "https://docs.example.com",
"description": "Complete guide to the platform"
}
]
}
// Action prompts in API response
{
"response": "I can help you with that...",
"prompts": [
{
"action_text": "Start Tutorial",
"action_url": "https://example.com/tutorial"
},
{
"action_text": "Contact Support",
"action_url": "mailto:support@example.com"
}
]
}
The HTML template can be adapted for React applications. See the React syntax section below for implementation details on using the HTML template for the chatbot in JSX components.
Are you sure you want to close the chat?
Your current conversation history will be permanently deleted.
Depending on your environment you'll want to use render
or include
. As a rule of thumb: server-side use include
, precompiled browser use render
. If you're using vf-eleventy you should use include
.
include
You'll need to pass a context object from your code or Yaml file (example), as well as the path to the Nunjucks template. Nunjucks' include
is an abstraction of render
and provides some additional portability.
{% set context fromYourYamlFile %}
- or -
{% set context = {
"component-type" : "element",
"hide-from-sitemenu" : true,
"config" : [object Object],
}
%}
{% include "../path_to/vf-chatbot-modal/vf-chatbot-modal.njk" %}
render
This approach is best for bare-bones Nunjucks environments, such as precompiled templates with the Nunjucks slim runtime where include
is not be available.
{% render '@vf-chatbot-modal', {
"component-type" : "element",
"hide-from-sitemenu" : true,
"config" : [object Object],
}
%}
<!-- Modal Chatbot -->
<div class="vf-content vf-chatbot-modal-container" data-vf-js-chatbot-modal-container data-vf-chatbot-config="{"type":"modal","title":"AI Assistant","welcome_logo":true,"welcome_message":"Welcome! I'm here to help","welcome_logo_alt":"AI Assistant","welcome_suggestions_title":"Try asking me:","input_placeholder":"Ask me ...","welcome_max_suggestions":4,"icons":{"assistant_avatar":"../../assets/vf-chatbot/assets/vf-chatbot--icon-16x16-dark-green.svg","user_avatar":"../../assets/vf-chatbot/assets/vf-chatbot--avatar-user.svg","send_button":"../../assets/vf-chatbot/assets/vf-chatbot--icon-send.svg","main_logo_url":"../../assets/vf-chatbot/assets/vf-chatbot--icon-32x32-dark-green.svg","minimize":"../../assets/vf-chatbot/assets/vf-chatbot--icon-minimize.svg","close":"../../assets/vf-chatbot/assets/vf-chatbot--icon-close.svg","selector_logo_url":"../../assets/vf-chatbot/assets/vf-chatbot--icon-24x24-dark-green.svg"},"api":{"chat_endpoint":"/api/chat","feedback_endpoint":"/api/feedback","qa_data_url":"../../assets/vf-chatbot/assets/vf-chatbot-qa.json","headers":{"Content-Type":"application/json","Authorization":"Bearer your-token"},"timeout":10000},"features":{"enable_welcome":true,"enable_feedback":true,"enable_sources":true,"enable_sources_custom_format":true,"enable_welcome_suggestions":true,"enable_typing_indicator":true,"enable_disclaimer":true,"enable_predefined_qa":true,"enable_fallback_responses":true,"enable_qa_data_loading":true,"enable_instant_feedback":false},"behavior":{"auto_scroll":true,"typing_delay":800,"show_scrollbar":false},"selectorContext":{"chatbotRoutes":{"multiSelect":true,"maxMultiSelect":3,"showSearch":true,"showSearchThreshold":5,"showAllServices":true,"showAllServicesSelected":true,"routes":"../../assets/vf-chatbot/assets/vf-chatbot-selector-services.json","placeholder":"Select services","title":"Services"}},"handlers":{"on_message_send":"handleMessageSend","on_response_receive":"handleResponseReceive","on_feedback_submit":"handleFeedbackSubmit","on_suggestion_click":"handleSuggestionClick","on_error":"handleError","on_conversation_start":"handleConversationStart","on_conversation_end":"handleConversationEnd"},"disclaimer":"Disclaimer: This chatbot is designed to assist you with general information and basic inquiries. See our <a class=\"vf-banner__link\" target=\"_blank\" rel=\"noopener noreferrer\" aria-label=\"disclaimer notes (opens in new tab)\" href=\"https://www.ebi.ac.uk/data-protection/privacy-notice/embl-ebi-public-website/\">disclaimer notes</a>.","footnote":"Review AI generated content for accuracy. <a class=\"vf-link\" target=\"_blank\" rel=\"noopener noreferrer\" aria-label=\"Leave feedback (opens in new tab)\" href=\"https://embl.service-now.com/esc?id=sc_cat_item&sys_id=5eeb8eb91b92e650b376da88b04bcbc1\">Leave feedback</a>."}">
<div class="vf-chatbot-modal__header">
<div class="vf-chatbot-modal__header-left">
<div class="vf-chatbot-selector" data-vf-js-chatbot-selector data-routes-path="../../assets/vf-chatbot/assets/vf-chatbot-selector-services.json" data-multiselect="true" data-max-multiselect="3" data-show-search="true" data-show-all-services="true" data-show-all-services-selected="true">
<button class="vf-chatbot-selector__title" data-vf-js-selector-toggle>
<img src="../../assets/vf-chatbot/assets/vf-chatbot--icon-24x24-dark-green.svg" alt="AI Assistant">
<div class="vf-chatbot-selector__title-content vf-u-margin__left--200">
<span class="vf-chatbot-selector__main-text">AI Assistant</span>
<span class="vf-chatbot-selector__title-text">Services</span>
</div>
<span class="vf-chatbot-selector__chevron">
<svg width="32" height="31" viewBox="0 0 32 31" fill="none" xmlns="http://www.w3.org/2000/svg">
<g clip-path="url(#clip0_3647_8230)">
<path d="M15.999 19.0975C15.7378 19.098 15.479 19.0468 15.2377 18.9468C14.9963 18.8469 14.7771 18.7001 14.5926 18.5151L8.32863 11.9279C8.21951 11.8137 8.13399 11.6791 8.07698 11.5318C8.01998 11.3845 7.99261 11.2274 7.99645 11.0695C8.00028 10.9116 8.03525 10.756 8.09934 10.6117C8.16342 10.4673 8.25537 10.337 8.36992 10.2283C8.48446 10.1195 8.61934 10.0344 8.76683 9.97791C8.91432 9.92139 9.07152 9.89454 9.2294 9.89889C9.38729 9.90325 9.54277 9.93872 9.68692 10.0033C9.83107 10.0678 9.96106 10.1602 10.0694 10.2751L15.7094 16.2143C15.7467 16.2537 15.7916 16.2851 15.8414 16.3066C15.8912 16.3281 15.9448 16.3391 15.999 16.3391C16.0533 16.3391 16.1069 16.3281 16.1567 16.3066C16.2065 16.2851 16.2514 16.2537 16.2886 16.2143L21.9286 10.2751C22.037 10.1602 22.167 10.0678 22.3112 10.0033C22.4553 9.93872 22.6108 9.90325 22.7687 9.89889C22.9266 9.89454 23.0838 9.92139 23.2312 9.97791C23.3787 10.0344 23.5136 10.1195 23.6282 10.2283C23.7427 10.337 23.8347 10.4673 23.8987 10.6117C23.9628 10.756 23.9978 10.9116 24.0016 11.0695C24.0055 11.2274 23.9781 11.3845 23.9211 11.5318C23.8641 11.6791 23.7786 11.8137 23.6694 11.9279L17.439 18.4991C17.2503 18.6888 17.0259 18.8394 16.7788 18.9421C16.5316 19.0448 16.2667 19.0976 15.999 19.0975Z" fill="#707372" />
</g>
<defs>
<clipPath id="clip0_3647_8230">
<rect width="16" height="16" fill="white" transform="translate(8 6.5)" />
</clipPath>
</defs>
</svg>
</span>
</button>
<div class="vf-chatbot-selector__dropdown" data-vf-js-selector-dropdown>
<div class="vf-chatbot-selector__search">
<label class="vf-u-sr-only" id="vf-chatbot-selector-search-label" for="vf-chatbot-selector-search">Type to search</label>
<input type="text" id="vf-chatbot-selector-search" aria-labelledby="vf-chatbot-selector-search-label" placeholder="Select services" data-vf-js-selector-search>
</div>
<div class="vf-chatbot-selector__header">
<span data-max-select="3">Select up to 3 services</span>
<a href="#" class="vf-chatbot-selector__clear" role="button" data-vf-js-selector-clear>Clear all</a>
</div>
<ul class="vf-chatbot-selector__list" data-vf-js-chatbot-selector-list>
<!-- Routes will be populated dynamically via JavaScript -->
</ul>
</div>
</div>
</div>
<div class="vf-chatbot-modal__header-right">
<button class="vf-chatbot-modal__minimize" aria-label="Minimize" data-vf-js-chatbot-modal-minimize>
<img src="../../assets/vf-chatbot/assets/vf-chatbot--icon-minimize.svg" alt="Minimize">
</button>
<button class="vf-chatbot-modal__close" aria-label="Close" data-vf-js-chatbot-modal-close>
<img src="../../assets/vf-chatbot/assets/vf-chatbot--icon-close.svg" alt="Close">
</button>
</div>
</div>
<div class="vf-chatbot-modal | vf-u-background-color-ui--grey--light" data-vf-js-chatbot-modal>
<div class="vf-chatbot-modal__content" data-vf-js-chatbot-modal-content>
<!-- Welcome Screen -->
<div class="vf-chatbot-welcome" data-vf-js-chatbot-welcome data-max-questions="4" data-enable-qa-data-loading="true" data-enable-predefined-qa="true" data-enable-fallback-responses="true" data-qa-data-url="">
<div class="vf-chatbot-welcome__content">
<div class="vf-chatbot-welcome__logo">
<img src="../../assets/vf-chatbot/assets/vf-chatbot--icon-32x32-dark-green.svg" alt="AI Assistant">
</div>
<h1 class="vf-chatbot-welcome__title">AI Assistant</h1>
<div class="vf-chatbot-welcome__message">
Welcome! I'm here to help
</div>
</div>
<div class="vf-chatbot-welcome__suggestions">
<p class="vf-chatbot-welcome__suggestions-title vf-u-margin__bottom--200">Try asking me:</p>
<div class="vf-chatbot-welcome__suggestions-grid" data-vf-js-chatbot-welcome-suggestions-grid>
<!-- Suggestions will be populated dynamically from qa.json using the template below -->
</div>
</div>
<!-- Template for welcome suggestions using vf-chatbot-action-prompt -->
<template id="welcome-suggestion-template">
<div class="vf-chatbot-action-prompt">
<a href="#" class="vf-chatbot-action-prompt__link vf-u-padding--200" role="button">
</a>
</div>
</template>
</div>
<!-- Messages Container -->
<div class="vf-chatbot-modal__messages-no-scrollbar" data-vf-js-chatbot-modal-messages data-auto-scroll="true">
<!-- Messages will be added here dynamically -->
</div>
<!-- Disclaimer Banner -->
<div class="vf-chatbot-modal__disclaimer" data-vf-js-chatbot-modal-disclaimer>
<div class="vf-banner vf-banner--alert vf-banner--info">
<div class="vf-banner__content">
<p class="vf-banner__text">Disclaimer: This chatbot is designed to assist you with general information and basic inquiries. See our <a class="vf-banner__link" target="_blank" rel="noopener noreferrer" aria-label="disclaimer notes (opens in new tab)" href="https://www.ebi.ac.uk/data-protection/privacy-notice/embl-ebi-public-website/">disclaimer notes</a>.</p>
<button role="button" aria-label="close notification banner" class="vf-button vf-button--icon vf-button--dismiss | vf-banner__button">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24">
<title>dismiss banner</title>
<path d="M14.3,12.179a.25.25,0,0,1,0-.354l9.263-9.262A1.5,1.5,0,0,0,21.439.442L12.177,9.7a.25.25,0,0,1-.354,0L2.561.442A1.5,1.5,0,0,0,.439,2.563L9.7,11.825a.25.25,0,0,1,0,.354L.439,21.442a1.5,1.5,0,0,0,2.122,2.121L11.823,14.3a.25.25,0,0,1,.354,0l9.262,9.263a1.5,1.5,0,0,0,2.122-2.121Z" />
</svg>
</button>
</div>
</div>
</div>
</div>
<!-- Input Container -->
<div class="vf-chatbot-modal__input-container vf-u-margin__top--400">
<div class="vf-chatbot-modal__input-wrapper">
<label class="vf-u-sr-only" id="vf-chatbot-modal-input-label" for="vf-chatbot-modal-input">Ask me</label>
<textarea id="vf-chatbot-modal-input" aria-labelledby="vf-chatbot-modal-input-label" class="vf-chatbot-modal__input vf-form__textarea vf-u-padding__left--400" placeholder="Ask me ..." rows="1" data-vf-js-chatbot-modal-input></textarea>
<button class="vf-chatbot-modal__send-button" aria-label="Send message" data-vf-js-chatbot-modal-send>
<img src="../../assets/vf-chatbot/assets/vf-chatbot--icon-send.svg" alt="Send">
</button>
</div>
<div class="vf-chatbot-modal__footnote vf-u-margin__top--200" data-vf-js-chatbot-modal-footnote>
Review AI generated content for accuracy. <a class="vf-link" target="_blank" rel="noopener noreferrer" aria-label="Leave feedback (opens in new tab)" href="https://embl.service-now.com/esc?id=sc_cat_item&sys_id=5eeb8eb91b92e650b376da88b04bcbc1">Leave feedback</a>.
</div>
</div>
<div class="vf-chatbot-dialog" data-vf-js-chatbot-dialog>
<div class="vf-chatbot-dialog__content">
<div class="vf-chatbot-dialog__header vf-u-margin__bottom--400">
<h2 class="vf-chatbot-dialog__title">Close chat and delete conversation?</h2>
<button class="vf-chatbot-dialog__close" data-vf-js-dialog-close aria-label="Close dialog">
<svg width="24" height="24" viewBox="0 0 24 24">
<path d="M19 6.41L17.59 5 12 10.59 6.41 5 5 6.41 10.59 12 5 17.59 6.41 19 12 13.41 17.59 19 19 17.59 13.41 12z" />
</svg>
</button>
</div>
<div class="vf-chatbot-dialog__body vf-u-margin__bottom--800">
<p>Are you sure you want to close the chat? <br>Your current conversation history will be permanently deleted.</p>
</div>
<div class="vf-chatbot-dialog__actions">
<button class="vf-chatbot-dialog__button vf-chatbot-dialog__button--outline" data-vf-js-dialog-cancel>
Keep chat open
</button>
<button class="vf-chatbot-dialog__button vf-chatbot-dialog__button--primary" data-vf-js-dialog-confirm>
Close and delete
</button>
</div>
</div>
</div>
<!-- Templates -->
<template id="feedback-positive-template">
<div class="vf-chatbot-feedback__form vf-u-margin__top--400">
<div class="vf-chatbot-feedback__form-content vf-u-padding--400">
<div class="vf-chatbot-feedback__form-content-header">
<div class="vf-chatbot-feedback__title">Tell us more (optional)</div>
<button role="button" class="vf-chatbot-feedback__form-close vf-button vf-button--icon vf-button--dismiss | vf-banner__button" type="button" aria-label="Close feedback form" data-vf-js-feedback-form-close>
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24">
<title>dismiss banner</title>
<path d="M14.3,12.179a.25.25,0,0,1,0-.354l9.263-9.262A1.5,1.5,0,0,0,21.439.442L12.177,9.7a.25.25,0,0,1-.354,0L2.561.442A1.5,1.5,0,0,0,.439,2.563L9.7,11.825a.25.25,0,0,1,0,.354L.439,21.442a1.5,1.5,0,0,0,2.122,2.121L11.823,14.3a.25.25,0,0,1,.354,0l9.262,9.263a1.5,1.5,0,0,0,2.122-2.121Z" />
</svg>
</button>
</div>
<div class="vf-chatbot-feedback__options">
<button class="vf-chatbot-feedback__option" data-feedback-option="accurate">
Accurate answer
</button>
<button class="vf-chatbot-feedback__option" data-feedback-option="easy">
Easy to understand
</button>
<button class="vf-chatbot-feedback__option" data-feedback-option="formatted">
Well formatted
</button>
<button class="vf-chatbot-feedback__option" data-feedback-option="helpful">
Helpful
</button>
</div>
<label id="vf-chatbot-feedback-comment-title" for=id="vf-chatbot-feedback-comment" class="vf-chatbot-feedback__comment-title">Comments</label>
<textarea id="vf-chatbot-feedback-comment" aria-labelledby="vf-chatbot-feedback-comment-title" class="vf-chatbot-feedback__comment" rows="4"></textarea>
<button type="submit" class="vf-chatbot-feedback__submit vf-u-padding--200" data-vf-js-feedback-submit>
Submit
</button>
</div>
</div>
</template>
<template id="feedback-negative-template">
<div class="vf-chatbot-feedback__form vf-u-margin__top--400">
<div class="vf-chatbot-feedback__form-content vf-u-padding--400">
<div class="vf-chatbot-feedback__form-content-header">
<div class="vf-chatbot-feedback__title">Tell us more (optional)</div>
<button role="button" class="vf-chatbot-feedback__form-close vf-button vf-button--icon vf-button--dismiss | vf-banner__button" type="button" aria-label="Close feedback form" data-vf-js-feedback-form-close>
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24">
<title>dismiss banner</title>
<path d="M14.3,12.179a.25.25,0,0,1,0-.354l9.263-9.262A1.5,1.5,0,0,0,21.439.442L12.177,9.7a.25.25,0,0,1-.354,0L2.561.442A1.5,1.5,0,0,0,.439,2.563L9.7,11.825a.25.25,0,0,1,0,.354L.439,21.442a1.5,1.5,0,0,0,2.122,2.121L11.823,14.3a.25.25,0,0,1,.354,0l9.262,9.263a1.5,1.5,0,0,0,2.122-2.121Z" />
</svg>
</button>
</div>
<div class="vf-chatbot-feedback__options">
<button class="vf-chatbot-feedback__option" data-feedback-option="inaccurate">
Inaccurate answer
</button>
<button class="vf-chatbot-feedback__option" data-feedback-option="nocontext">
Did not use context
</button>
<button class="vf-chatbot-feedback__option" data-feedback-option="poorformat">
Poorly formatted
</button>
<button class="vf-chatbot-feedback__option" data-feedback-option="nothelpful">
Not helpful
</button>
</div>
<label id="vf-chatbot-feedback-comment-title" for=id="vf-chatbot-feedback-comment" class="vf-chatbot-feedback__comment-title">Comments</label>
<textarea id="vf-chatbot-feedback-comment" aria-labelledby="vf-chatbot-feedback-comment-title" class="vf-chatbot-feedback__comment" rows="4"></textarea>
<button type="submit" class="vf-chatbot-feedback__submit vf-u-padding--200" data-vf-js-feedback-submit>
Submit
</button>
</div>
</div>
</template>
<template id="user-message-template">
<div class="vf-chatbot-message vf-chatbot-message--user vf-u-margin__top--400">
<div class="vf-chatbot-message__avatar vf-u-margin__bottom--200">
<span class="vf-chatbot-message__avatar-name">You</span>
<img src="../../assets/vf-chatbot/assets/vf-chatbot--avatar-user.svg" alt="You">
</div>
<div class="vf-chatbot-message__content vf-u-padding--200">
<div class="vf-chatbot-message__content-prompt vf-u-padding__left--200 vf-u-padding__right--200">
Hello!
</div>
</div>
</div>
</template>
<template id="assistant-message-template">
<div class="vf-chatbot-message vf-chatbot-message--assistant vf-u-margin__top--400">
<div class="vf-chatbot-message__avatar vf-u-margin__bottom--200">
<img src="../../assets/vf-chatbot/assets/vf-chatbot--icon-16x16-dark-green.svg" alt="AI Assistant">
<span class="vf-chatbot-message__avatar-name">AI Assistant</span>
</div>
<div class="vf-chatbot-message__content vf-u-padding--200">
<div class="vf-chatbot-message__content-prompt vf-u-padding__left--200 vf-u-padding__right--200">
How can I help you?
</div>
</div>
</div>
<div class="vf-chatbot-feedback vf-u-margin__top--200" data-vf-js-chatbot-feedback></div>
</template>
<template id="loading-indicator-template">
<div class="vf-chatbot-message vf-chatbot-message--assistant vf-chatbot-message--loading vf-u-margin__top--400">
<div class="vf-chatbot-message__avatar vf-u-margin__bottom--200">
<img src="../../assets/vf-chatbot/assets/vf-chatbot--icon-16x16-dark-green.svg" alt="AI Assistant">
<span class="vf-chatbot-message__avatar-name">AI Assistant</span>
</div>
<div class="vf-chatbot-message__content vf-u-padding--200">
<div class="vf-chatbot-message__content-loading-dots">
...
</div>
</div>
</div>
</template>
<template id="action-prompts-template">
<div class="vf-chatbot-action-prompts">
<div class="vf-chatbot-action-prompts__list" data-vf-js-action-prompts-list>
<!-- Individual prompts will be populated here -->
</div>
</div>
</template>
<template id="single-action-prompt-template">
<div class="vf-chatbot-action-prompt">
<a href="#" class="vf-chatbot-action-prompt__link vf-u-padding--200" role="button">
</a>
</div>
</template>
</div>
</div>
This repository is distributed with npm. After installing npm and yarn, you can install vf-chatbot-modal
and other dependant components with this command.
yarn add @visual-framework/vf-chatbot @visual-framework/vf-chatbot-modal @visual-framework/vf-chatbot-fab @visual-framework/vf-chatbot-dialog @visual-framework/vf-chatbot-action-prompt @visual-framework/vf-chatbot-feedback @visual-framework/vf-chatbot-prompt @visual-framework/vf-chatbot-selector @visual-framework/vf-chatbot-sources @visual-framework/vf-chatbot-welcome
@import "@visual-framework/vf-chatbot-modal/index.scss";
import { initVFChatbot } from 'vf-chatbot/vf-chatbot.js';
import { initVFChatbotModal } from 'vf-chatbot-modal/vf-chatbot-modal.js';
import { initVFChatbotFab } from 'vf-chatbot-fab/vf-chatbot-fab.js';
window.addEventListener("load", () => {
initVFChatbot(config);
});
where config is the configuration object with different options as described below
const config = {
// Basic Settings
type: "modal",
chatbotBottomMargin: 100, //This is optional and is used to apply bottom margin in pixels
title: "AI Assistant",
welcome_message: "Welcome! I'm here to help",
input_placeholder: "Ask me anything...",
welcome_max_suggestions: 4,
// Content & Branding
disclaimer: 'Custom disclaimer with <a href="/privacy">privacy policy</a>',
footnote: 'Custom footnote with <a href="/feedback">feedback link</a>',
// Icons & Assets
icons: {
assistant_avatar: "path/to/assistant-icon.svg",
user_avatar: "path/to/user-icon.svg",
send_button: "path/to/send-icon.svg",
main_logo_url: "path/to/logo.svg",
selector_logo_url: "path/to/selector-logo.svg",
},
// API Configuration
api: {
chat_endpoint: "/api/chat",
feedback_endpoint: "/api/feedback",
qa_data_url: "path/to/qa-data.json",
headers: {
"Content-Type": "application/json",
"Authorization": "Bearer your-token"
},
timeout: 10000
},
// Feature Toggles
features: {
enable_welcome: true,
enable_feedback: true,
enable_sources: true,
enable_sources_custom_format: true,
enable_welcome_suggestions: true,
enable_typing_indicator: true,
enable_disclaimer: true,
enable_predefined_qa: true,
enable_fallback_responses: true,
enable_qa_data_loading: true,
enable_instant_feedback: false
},
// New options
behavior: {
auto_scroll: true,
typing_delay: 800,
show_scrollbar: false // or false to hide scrollbar
},
selectorContext: {
chatbotRoutes: {
multiSelect: false,
showSearch: true,
showSearchThreshold: 5,
routes: "assets/vf-chatbot-selector-services.json",
placeholder: "Search",
title: "Services"
}
},
handlers: {
on_message_send: "handleMessageSend",
on_response_receive: "handleResponseReceive",
on_feedback_submit: "handleFeedbackSubmit",
on_suggestion_click: "handleSuggestionClick",
on_error: "handleError",
on_conversation_start: "handleConversationStart",
on_conversation_end: "handleConversationEnd"
}
};
The chatbot includes an integrated selector component which can be configured to present different selection options to user:
const selectorConfig = {
selectorContext: {
chatbotRoutes: {
// Multi-selection settings
multiSelect: true,
maxMultiSelect: 3,
// Search functionality
showSearch: true,
showSearchThreshold: 5,
// "All Services" option
showAllServices: true,
showAllServicesSelected: true,
// Data source
routes: "assets/vf-chatbot-selector-services.json",
// UI labels
placeholder: "Search",
title: "Available Services"
}
}
};
Selector Data Format (JSON):
{
"routes": [
{
"id": "service-1",
"title": "Service 1 title"
},
{
"id": "service-2",
"title": "Service 2 title"
}
]
}
Chatbot comes with a provision to allow custom event handlers. These handlers can be defined in your code to handle specific actions for different events triggered during interaction with chatbot. Configure custom handlers for chatbot events:
const config = {
handlers: {
on_message_send: "handleMessageSend",
on_response_receive: "handleResponseReceive",
on_feedback_submit: "handleFeedbackSubmit",
on_suggestion_click: "handleSuggestionClick",
on_error: "handleError",
on_conversation_start: "handleConversationStart",
on_conversation_end: "handleConversationEnd"
}
};
// Implement handler functions
function handleMessageSend(message, conversationId) {
console.log('User sent:', message);
// Track analytics, log conversations, etc.
}
function handleResponseReceive(response, sources, prompts) {
console.log('Assistant responded:', response);
// Process response, update UI, etc.
}
function handleFeedbackSubmit(feedbackData) {
console.log('Feedback received:', feedbackData);
// Send to analytics, update models, etc.
}
Likewise you can also listen to events emitted by the chatbot for specific interactions:
// Listen for specific chatbot events
document.addEventListener('vf-chatbot:message-send', (event) => {
const { message, conversationId } = event.detail;
// Handle message send
});
document.addEventListener('vf-chatbot:message-receive', (event) => {
const { message, conversationId } = event.detail;
// Handle message send
});
document.addEventListener('vf-chatbot-feedback:submit', (event) => {
const { messageId, feedbackType, feedbackText, feedbackComment } = event.detail;
// Handle feedback submission
});
document.addEventListener('vf-chatbot-welcome:suggestion-click', (event) => {
const { question } = event.detail;
// Handle suggestion clicks
});
document.addEventListener('vf-chatbot:assistant-change', (obj) => {
const { selectedRoutes } = obj.selectedAssistants;
const { conversationId } = obj.conversationId;
// Handle service selection
});
document.addEventListener('vf-chatbot:error', (event) => {
const { message, conversationId } = event.detail;
// Handle message send
});
document.addEventListener('vf-chatbot:conversation-start', (event) => {
const { message, conversationId } = event.detail;
// Handle conversation start
});
document.addEventListener('vf-chatbot:conversation-end', (event) => {
const { message, conversationId } = event.detail;
// Handle conversation end
});
const config = {
features: {
enable_feedback: true,
enable_instant_feedback: false // Default
}
};
const config = {
features: {
enable_feedback: true,
enable_instant_feedback: true // Thumbs up/down only
}
};
Load predefined questions and answers:
const config = {
features: {
enable_predefined_qa: true,
enable_qa_data_loading: true
},
api: {
qa_data_url: "path/to/qa-data.json"
}
};
Q&A Data Format:
{
"predefinedQA": {
"How can I submit genomic data to EMBL-EBI?": {
"answer": "To submit genomic data, visit the EMBL-EBI submission portal, where you’ll find step-by-step guides and tools for submitting sequencing data, assemblies, annotations, and more.",
"sources": [
{
"domain": "ebi.ac.uk",
"title": "EMBL's European Bioinformatics Institute",
"url": "https://www.ebi.ac.uk/",
"description": "Run BLAST searches against comprehensive sequence databases at EMBL-EBI."
},
{
"domain": "ena-docs.readthedocs.io",
"title": "ENA Documentation",
"url": "https://ena-docs.readthedocs.io/en/latest/",
"description": "ENA Documentation"
}
]
}
},
"fallbackResponses": [
{
"answer": "I'm sorry, I'm having trouble connecting to my knowledge base right now. Could you try again in a moment?",
"prompts": [
{
"action_text": "Contact support team",
"action_url": "tel:+44 1223 494 444"
},
{
"action_text": "Submit a support request",
"action_url": "https://www.ebi.ac.uk/about/contact/support/"
}
]
}
]
}
Your chat API should accept POST requests:
// Request format
{
"message": "User's question",
"conversationId": "unique-id",
"context": {
"selectedServices": ["service-1", "service-2"]
}
}
// Response format
{
"response": "Assistant's answer",
"sources": [
{
"title": "Documentation Link",
"url": "https://example.com/docs"
}
],
"prompts": [
{
"action_text": "Learn More",
"action_url": "https://example.com/learn"
}
]
}
const config = {
features: {
enable_welcome: true,
enable_welcome_suggestions: true
},
welcome_logo: true,
welcome_message: "Welcome to our AI assistant!",
welcome_suggestions_title: "Popular questions:",
welcome_max_suggestions: 6
};
const config = {
features: {
enable_sources: true
enable_sources_custom_format: true,
}
};
// Sources in API response
{
"response": "Here's the information...",
"sources": [
{
"title": "Official Documentation",
"url": "https://docs.example.com",
"description": "Complete guide to the platform"
}
]
}
// Action prompts in API response
{
"response": "I can help you with that...",
"prompts": [
{
"action_text": "Start Tutorial",
"action_url": "https://example.com/tutorial"
},
{
"action_text": "Contact Support",
"action_url": "mailto:support@example.com"
}
]
}
The HTML template can be adapted for React applications. See the React syntax section below for implementation details on using the HTML template for the chatbot in JSX components.
File system location: components/vf-chatbot-modal
Find an issue on this page? Propose a change or discuss it.
Are you sure you want to close the chat?
Your current conversation history will be permanently deleted.