How to Use AI Agent in Bit Flows | Build Agentic Workflow in WordPress

AI Agent is a smart system that can make decisions and take actions on its own, based on the information it receives. It’s like a virtual assistant that can understand instructions, analyze data, and perform tasks automatically, without needing constant human input.

AI Agent in Bit Flows brings intelligent decision-making capabilities into WordPress automation workflows. Instead of following only a fixed set of steps, an AI Agent can understand data, decide what to do next, and change its actions based on different situations.

Use the AI Agent in The Best WordPress Automation Plugin – Bit Flows

Check out our easy-to-follow tutorial!

This guide will show you how to set up and use the AI Agent in WordPress → Bit flows to create smarter and more flexible automation workflows.

Step 1: Add the Trigger

First, activate Bit Flows, then go to your WordPress Dashboard → Bit Flows → Flows → Create Flow, and give your flow a meaningful name. Follow this documentation to learn how to create a workflow.

Start by selecting a trigger to capture incoming support requests. This could be a Form Submission trigger, a Custom Webhook (if your form can send webhook data), or a WooCommerce New Order trigger. Configure the trigger so it runs whenever a form is submitted or a new order is placed.

For example, select Bit Form as the trigger and choose your preferred form. Then click the “Listen Response” button and submit a test form. This allows Bit Flows to capture the form data, which will be used in the next steps of the workflow.

Step 2: Add an AI Agent Node

Click the (+) icon in the Flow Builder and open the Tools section. Find the “AI Agent” and click on it. You can also add the AI Agent from the Bit Flows Tools tab.

After adding the AI Agent to your workflow, you’ll see the interface of the AI Agent. This is where you tell the Agent what task to perform. Enter your prompt, system prompt, and set the maximum iterations to control how many times the agent can process the task.

Step 03: Set Up the LLM Chain 

Click on the AI Agent block and add a Chat Model like OpenAI or DeepSeek. Then connect it with your API key. And then select your preferred model. You can also enable the Show Advanced Features to add token limits, temperature, and more.

Step 04: Configure Simple Memory (Optional)

If you want the AI Agent to remember information from previous steps, such as a customer’s name or a ticket number, enable the Simple Memory node. You can store variables like {customer_name} or {ticket_number}, which can then be referenced in later prompts or actions.

Step 05: Add Tool/Action Calls

Under the tools of the AI Agent node, you can use any of the Bit Flows actions. In this example, the data will be routed to three different platforms. First, the AI Agent will analyze the form data to understand the user’s intent and sentiment, then execute actions based on the prompts you have configured.

After adding all the tools, select the event type you want to use.

Step 06: Add Prompts to AI Agent and Tools

Now it’s time to instruct the AI Agent by adding a prompt. Click on the AI Agent node in Bit Flows, add your preferred prompts, and set the maximum iteration limits. Once the prompt is added, you can run a test by clicking the ‘Test Run” button to ensure everything works as expected.

You can also add a short description for each tool. For example, I’ve added a brief note to the Telegram tool.

Step 07: Let the AI Agent Define the Result

Now comes the exciting part, letting the AI Agent determine the results automatically instead of mapping fields manually. Click the AI Define icon beside the field, and the Bit Flows AI Agent will automatically decide what data to send to that field.

Define each field using the AI Agent’s responses according to your prompts and requirements.

You can also add actions after the AI Agent node. For example, I’ve connected a Google Sheet to track every form submission. Here is my final workflow :

From now on, each time a form is submitted, the Bit Flows AI Agent will analyze the information, detect the user’s intent and sentiment, and determine the next actions automatically. 

To test this, I submitted a sample form with the following information.

Here are the execution logs of the flow, which show each step in detail, including the AI Agent’s prompt and its corresponding response. You can review these logs to see how the agent interpreted the data and adjust your prompts, memory settings, or routing as needed.

This log shows how the AI Agent made the decision. It starts by reviewing the message, then uses the WooCommerce tool to check the order status. After that, it sends a Telegram notification and emails a quick reply to the customer. 

Here’s the output from Bit Flows AI Agent :

From now on, every incoming support request will be reviewed by the AI Agent first. You and the customer will get an instant notification.

This is just a basic example. The use cases for any agent flow are similar: select a trigger, add the AI Agent node, set up the LLM and memory, write your prompt, and connect tools to perform actions.