What are AI Chat Agents?

AI Chat Agents are Regal AI agents that handle real-time conversations with customers through a web chat experience.

Regal's AI Chat widget, powered by the Regal AI Agent, is easily embeddable in your website (via JavaScript). When a customer sends a message, the AI Agent responds instantly and can guide them toward resolution, conversion, or escalation as needed. Customers can interact with the AI Agent via whatever channel they prefer - whether it be chat, webRTC voice (enabled via the "Enable Voice" button), or both.

💬

Try it yourself - open the widget in the bottom right corner of this page to chat with Reggie.

Common AI Chat Use Cases

• Customer Support & Troubleshooting
  • Resolve customer issues quickly with dynamic troubleshooting directly from your website.
• Onboarding & Setup Guidance
  • Guide new users through setup and feature discovery as they navigate your product, reducing friction and improving activation.
• Appointment Scheduling & Follow-ups
  • Collect required information and assist customers with booking demos, consultations, or support callbacks.
• After-Hours Coverage
  • Handle inbound conversations outside business hours so customers always receive an immediate response.
• Lead Qualification & Sales Intake
  • Engage inbound visitors, ask qualifying questions, and route high-intent leads to sales or booking flows in real time.

Deploying AI Chat Agents

1. Create your SDK instance

1. Share the whitelist of origins (domains where the widget will run) with your Regal Forward Deployed Engineer.
2. Regal will create your SDK instance and return your SDK Instance ID, which you’ll use during widget initialization.

2. Add the script tag to your header

Add the Regal widget SDK script tag inside the header of your website:

<script src="https://cdn.regal.io/conversations-sdk/production/regal-widget.umd.js"></script>

3. Initialize the widget in the body

At the end of your body, add the following script tag to initialize the widget with your brand and agent details.

<script>
  window.RegalWidget.init({
    apiUrl: 'https://api.regal.ai',
    brandSlug: '...your brand slug...',
    agentId: '...the UUID of your agent provided by Regal...',
    widgetHeaderText: '...your desired widget header (Chat with Regal)...',
    sdkInstanceId: '...the SDK instance ID provided by Regal...',
  });
</script>

Once initialized, chat messages are automatically routed to the specified AI Chat Agent.

4. 🚀 You're live!

Once deployed, visitors can immediately start chatting with your AI Agent, and conversations will flow through Regal just like any other channel. From here, you can iterate by:

• Monitoring conversations and outcomes in Regal Reporting
• Refining agent instructions and prompting
• Enhancing knowledge base coverage

Your AI Chat Agent is ready to meet customers where they want to be met - quickly and easily.

Enhanced Configurability

Identify the Contact for More Personalized Interactions

If you know the user's phone number or email, set it at any point after the widget is initialized and before the contact sends their first message. Passing contact identifiers allows Regal to associate the interaction with the existing contact profile and enables the AI agent to reference and update customer information, allowing for more personalized interactions.

`window.RegalWidget.updateContactPhone("+19175550000");
window.RegalWidget.updateContactEmail("[email protected]");`

Dynamically Update Agent Powering the Widget

If you want to change which AI agent handles the conversation between interactions, update the agent ID dynamically:

`window.RegalWidget.updateAgentId('...agent UUID...');`

This is commonly used when routing different pages or user states to different agents.