Pizza Ordering Chatbot using AI Agents and Tools - n8n Workflow

Restaurants or food service providers looking to automate order intake.
Businesses needing customer service automation using AI agents.
Developers exploring advanced Langchain agent capabilities and custom tools within an n8n workflow.
Users seeking complex, practical n8n templates for conversational AI.

This comprehensive n8n workflow demonstrates how to build a smart, conversational assistant named 'Pizzaro' capable of managing a digital pizza ordering process. By utilizing the specialized Langchain agent n8n node, this automation interprets natural language requests and intelligently delegates tasks to specific tools (webhooks) for backend operations. This robust system handles everything from fetching menu details to processing and tracking orders, minimizing human intervention. This advanced n8n workflow is a prime example of leveraging external services and powerful AI to create scalable customer service solutions, providing high-value n8n templates for complex integrations.

The entire process of this n8n workflow is driven by a conversational AI loop, initiated by a specialized n8n trigger.

1. Trigger and Greeting: The automation starts with the When chat message received n8n trigger node. This immediately presents the customer with a welcome message from 'Pizzaro' and waits for input.


2. AI Orchestration: The user's input is passed directly to the central AI Agent n8n node. This agent is configured with Chat OpenAI as its language model and uses the Window Buffer Memory n8n node to maintain context throughout the conversation.


3. Tool Selection: Based on the user's intent, as defined in its detailed system prompt, the AI Agent dynamically selects the appropriate tool:

1. Response Generation: The agent executes the chosen tool, receives the operational data (e.g., menu list, order confirmation), and uses the Langchain n8n node to synthesize a natural, coherent response delivered back to the customer, completing the cycle of this powerful n8n workflow.

To set up this n8n workflow, follow these steps:

1. Import: Copy the entire JSON code provided and import it into your n8n instance.


2. OpenAI Credentials: Locate the Chat OpenAI n8n node and update the 'OpenAI Credentials' with your valid API key.


3. Agent Configuration: Review the AI Agent n8n node. The system message defines the agent's persona and rules. Adjust these if your menu or ordering process changes.


4. Tool Webhooks: The custom tools (Get Products, Order Product, Get Order) are currently set to placeholder n8n URLs (https://n8n.io/webhook/...). You must replace these URLs with the actual webhooks or API endpoints that manage your product catalog, order processing, and status tracking in your backend system.


5. Activate: Once credentials and webhook URLs are configured, activate the n8n workflow. The When chat message received n8n trigger is now ready to handle incoming chat sessions.

When chat message received (n8n trigger): The starting point of this n8n workflow. It acts as the chat interface trigger, initiating the automation upon receiving a user message. It also handles the initial welcome message.
AI Agent (n8n node): The core intelligence. It coordinates the conversation, selects tools based on intent, and manages the flow. Key configuration includes a comprehensive system message defining the agent's ordering and status rules.
Chat OpenAI (n8n node): Serves as the Large Language Model provider for the AI Agent, handling interpretation and response generation.
Window Buffer Memory (n8n node): Crucial for conversational flow. This Langchain n8n node allows the AI Agent to remember previous messages, ensuring context is maintained throughout the customer's interaction.
Get Products (n8n node / ToolHttpRequest): Configured as a tool that the AI Agent can call. Its function is to retrieve and present the available pizza menu using a defined webhook URL. The agent knows to use this tool when asked about the menu.
Order Product (n8n node / ToolHttpRequest): This tool processes the actual order. It's configured as a POST request, sending the details parsed by the AI Agent (e.g., pizza type, quantity) to a backend webhook to place the order.

• Get Order (n8n node / ToolHttpRequest): A dedicated tool for checking the status of previous orders, used by the AI Agent when a customer inquires about tracking.