ChatDaddy
Automation & Message Flows

How to use the AI Chatbot node

- Navigate to Automation > Message Flows. - Open or create a new flow - Click the “Choose next step” button after a trigger, condition, or message button to add a new node. - Selec…

Updated Apr 15, 2026

💡 Note: The AI Chatbot node allows you to connect smart chat agents to your automated message flows. These agents can qualify leads, answer FAQs, or handle customer conversations—all with AI.

🧭 How to add AI Chatbot node

Note: AI Chatbot node node can be used in after buttons, conditions or triggers

1. Open Your Message Flow

  • Navigate to Automation > Message Flows.

  • Open or create a new flow

2. Add the AI Chatbot Node

  • Click the “Choose next step” button after a trigger, condition, or message button to add a new node.

  • Select “AI Chatbot” from the list of available apps.

3. Choose an AI Chat Agent

  • Under the Integration section, click the dropdown to select an existing AI chat agent

✅ Tip: If you haven’t created a bot yet, click the ➕ button beside the dropdown to create a new AI agent.

You can click here to learn how to create one

  • You’ll see a list of your AI agents (with their creation dates for easy reference).

  • Select one to link it to this step.

4. Configure App Inputs

  • This should contain the question you want to ask the AI Chatbot. You can either:

    • Pass a static question like: “Send the pricing details of Product A” OR

    • Use the {{text}} variable from the incoming message trigger to enable the AI Chatbot to have a conversation with your client

5. Add a Condition Node

After the AI Chatbot processes the message, you can check whether the bot successfully handled it.

  • Add a Condition node.

  • Create two branches using the AI’s response status:

    • If “Successful Response” is true → means the bot gave a valid answer. You can go ahead and connect this to a message node with the {{reply}} variable to send the bot’s reply

    • If “Successful Response” is false → means the bot didn’t have enough information or there was some error while generating the response. In such cases, we can ddd a Message node with a fallback message like this and follow it with an Action node to assign the ticket to a human agent

“I’m sorry, I do not have the data to answer your question. Please allow me to connect you to a live agent so they can help you with this.”

🧪 Example Use Cases

  • 🤖 Qualify leads before assigning to sales

  • 💬 Answer product questions

  • 🌐 Switch language based on customer input

  • 🧠 Helpdesk support before creating tickets

Feedback
Was this article helpful?

Related articles