Skip to main content
Rich Responses allow AI Agents in ChatbotX to automatically send visual interactive messages such as product cards (Card), horizontal collections (Carousel), quick selection buttons (Quick Replies, Buttons), images, videos, and documents instead of relying on plain text alone. At the same time, AI Agents can directly execute management actions such as assigning Tags, updating Custom Fields, triggering automated Flows, or transferring conversations to live support agents. Typically, standard AI models can only reply with raw text paragraphs. If a business wants to send product images or price tables, standard AI only provides external links, forcing customers to click away to another website. With Rich Responses, ChatbotX acts as a middleware layer that automatically translates AI responses into standardized interactive message formats tailored to each conversation channel (Messenger, Instagram, WhatsApp, Zalo, Webchat).

When to Use Rich Responses

Using Rich Responses delivers a professional customer experience and shortens the buyer conversion journey in key scenarios:
  • When consulting on products or services that require displaying images, names, prices, and direct purchase buttons.
  • When suggesting quick choices (Quick Replies) to optimize typing effort for mobile users.
  • When collecting customer information (Email, Phone number, Full name) and automatically saving it into Custom Fields.
  • When automatically segmenting customers (assigning Tags like lead, hot-lead, vip, support-needed) based on conversation flow.
  • When triggering advanced automated scenarios (Flows) or transferring chats to live agents for special requests.

How to Set Up Rich Responses for AI Agents

1

Create Custom Fields and Tags in Settings

Go to ChatbotX System Settings and create the required Custom Fields (e.g., size, color, address, interest, buyer_status) and Tags (e.g., lead, hot-lead, support-needed) that your business needs.
2

Access AI Agents Management

From the left navigation bar, select AI Agents.
3

Select Agent to Configure

Click edit on an existing Agent or create a new Agent you want to configure with Rich Responses.
4

Enable Rich Response Switch

In the Agent configuration modal, scroll down to the Rich Response toggle and switch it to ON. This setting enables the Agent to output a structured JSON response containing messages and actions instead of raw plain text streaming.
Enable Rich Response
5

Paste Prompt Template into System Message

Copy the production-ready Prompt configuration template below and paste it into the System Message (Prompt) field of your Agent.
6

Save Settings and Test

Click Confirm (or Save Settings) and test-chat with the Agent on a live conversation channel to verify visual rendering.
Important Note on Custom Fields: For an AI Agent to save data into a Custom Field, you must create that Custom Field in ChatbotX System Settings first. If the data field has not been created beforehand, the Agent cannot record or save values to the system.

Production Prompt Template for AI Agents

Below is a complete production-grade Prompt template designed for automated sales and customer support AI Agents. This prompt fully integrates Rich Responses output rules (Service Cards, Carousels, size/color Quick Replies) and automatically executes system Actions (assigning Tags, saving Custom Fields, running Flows, transferring to Human Agents). You can copy the entire template below for direct use:

Best Practices and Important Notes

Rich Message formats (Card, Carousel, Quick Reply) are automatically optimized and rendered by ChatbotX for each specific conversation channel. For example: On Messenger, they display as standard Facebook Carousels, while on Zalo or WhatsApp, they render as compatible list views with visual images.
When configuring the JSON array in your System Message, make sure that the Flow ID, Tag names, and Custom Field names match exactly what you created in ChatbotX System Settings.