For the complete documentation index, see llms.txt. This page is also available as Markdown.

Quick Reply

Create a Quick Reply button row that an Agent can attach to its responses to offer clickable options and guide users to the next step

Quick Reply is a type of Agent UI. After an Agent responds, it adds a row of clickable buttons (chips) below the message so users can choose their next step without typing: send a message, open a link, or prefill the input field. The Agent uses the "Trigger Conditions" to decide when to display the buttons, with up to 13 buttons per row.

When to use it

A telecom support team wants the Agent to ask "How can I help you?" and show three buttons—"Check bill," "Change plan," and "Report an issue"—when a user starts a conversation without a clear intent. When the user clicks "Check bill," the Agent goes straight to the billing flow. Another use is dynamic generation: when a user discusses a product they are interested in, the LLM generates buttons for up to 5 related products based on the context.

The difference from Quick Menu: a quick menu stays in the input bar and does not change with the conversation. Quick Reply is attached below a specific response; the Agent decides when it appears, and its content can be generated in real time based on the conversation.

Create a Quick Reply

1

Select a type

Go to "AI Features" → "Agent UI," click "Add Agent UI," and select Quick Reply (a quick reply button row).

Quick Reply card on the Agent UI type selection page
Click Quick Reply under "Select Agent UI Type"
2

Enter basic information

Field
Description

Display Name

The name displayed in the admin console, such as "Support Menu"

Tool Name

An English identifier used by the LLM; it must follow the naming rules (English letters and underscores, with no spaces)

Description

Optional; describes what this set of buttons is for

Trigger Conditions

Required. Tell the Agent in natural language when to attach this set of buttons, such as "When a user starts a conversation without a clear intent and needs help choosing" or "Provide category shortcuts when users ask frequently asked questions"

The more specific the trigger conditions, the better the Agent can display buttons at the right time instead of attaching them to every response. When attaching multiple Quick Reply sets to the same Agent, avoid overlapping trigger conditions.

3

Select a template (optional)

The "Select Template" section lists built-in templates. Click one to display a preview on the right. Clicking "Apply This Template" overwrites the current button settings with the template content, which you can then edit.

Quick Reply editor and preview after applying the Support Menu template
After applying the "Support Menu" template, the preview on the right simulates the button row in Web Chat
4

Configure buttons

First, select Static List or Dynamic Generation above "Button Settings" (see the next section for the differences), then configure each button. The "Preview" on the right simulates the Web Chat display in real time. Click a button to try it, then click "Show Buttons Again" to start over.

5

Save

Check the preview, then click "Save" in the upper-right corner. Before saving, the system checks that button text is not empty, fixed text is no longer than 20 characters, AI fields have prompts, links start with https://, and the button count range is within 1–13.

Static list and dynamic generation

Mode
How buttons are determined
Best for

Static List

Configure 1–13 buttons yourself, choosing "Fixed" or "AI" for each button's text and action fields

Predefined service categories and FAQ shortcuts

Dynamic Generation

Provide a "button pattern" and a count range; the LLM generates N buttons in real time based on the conversation context

Product recommendations and options based on topics users mention

Static list

Click "Add Button" to add a button (up to 13). Expand each button to configure the fields described in the next section; use the trash icon to the right of its title to delete it. Each button's title shows its action type label and button text for easy identification.

Dynamic generation

Field
Description

Minimum Count/Maximum Count

The range of buttons the LLM generates; must satisfy 1 ≤ minimum ≤ maximum ≤ 13

Generation Rules

Tell the LLM when and how to generate buttons, and how many, such as "List up to 5 related products based on the user's interests; skip the entire set if there are no relevant items"

Button Pattern

Settings shared by all generated buttons (action type, text, and icon); the text field is usually set to AI generation

Dynamic generation mode does not provide JSON source editing.

Settings for each button

Action types

Action
When clicked
Fields to fill in

Send Message

Sends a message to the Agent as the user

Message to Send: the actual text sent, which can differ from the button text

Open Link

Opens a URL

Link URL: must start with https://

Prefill Input Field

Inserts text into the input field without sending it; users can edit it before sending

Prefill Text

Fixed and AI

"Button Text" and the action fields in the table above each have a Fixed/AI toggle on the right:

  • Fixed: Displays the value you enter. Button text can be up to 20 characters.

  • AI: Enter a prompt, and the LLM generates content from it during the conversation. For example, set button text to AI and enter "The product name the user just mentioned."

Button icons

Each button can use None, Image (enter an https:// image URL), Emoji (a single emoji), or Initials Badge (1–2 characters).

Advanced action JSON (LINE only)

To use native LINE actions beyond the three basic actions in LINE (such as datetimepicker, postback, camera, cameraRoll, location, richmenuswitch, or clipboard), click an example label at the top to insert it quickly, or paste a LINE action object as JSON. The system passes it to LINE unchanged. These buttons appear gray in the preview and are not displayed in Web Chat.

JSON source

The "JSON Source" section (collapsed by default) corresponds to the LINE quickReply JSON for the current static list. Advanced users can edit it directly and click "Apply JSON to Button Settings." Applying it overwrites the button settings and only supports the message/uri/inputOption actions; any buttons beyond 13 are discarded.

Attach to an Agent

After saving, attach it to an Agent that uses Agent Mode for it to take effect: go to the Agent settings → "Interface & Experience" → "Agent UI" → "Select Agent UI," select the Quick Reply you just created, and save. For detailed instructions, see Bind to an Agent.

Frequently asked questions

Which platforms display the buttons?

Web Chat and LINE. Both support the three basic actions; "Advanced action JSON" only works in LINE.

Why didn't the Agent attach any buttons?

First, check that the Agent's response mode is Agent Mode, that this Quick Reply set is selected on the Agent UI tab, and that the situation described in "Trigger Conditions" matches the user's actual question. In dynamic generation mode, if the LLM determines that there are no suitable items based on the "Generation Rules," it skips the entire set.

Do the buttons disappear after the user clicks one?

Yes. The buttons are only attached below that response. Once one is clicked, the entire row collapses (the preview displays "Buttons Hidden"). The Agent decides again whether to attach buttons to its next response.

Last updated

Was this helpful?