Turn a simple chat window into a live order desk. Customers can ask for the menu, place orders, and check order status in one place. Ideal for food shops and small retail teams that want faster order intake without extra staff.

A public chat trigger collects the message and passes it to an AI agent powered by OpenAI. The agent uses short term memory to keep the conversation on track. It can call three HTTP tools to get the menu, submit an order, and fetch order status. A calculator tool helps with quick math checks like totals or quantities. The flow routes every request through the model, which decides which tool to call, then returns a clear answer back to the user. This end to end chat flow reduces back and forth and speeds up checkout.

Setup is simple. Add your OpenAI API key, point the HTTP tools to your own endpoints, and adjust the system message for your products and rules. Expect shorter wait times, fewer mistakes in orders, and more chats handled at once. Useful for food ordering, quick retail purchases, or booking simple services with minor changes.

• Public chat trigger that greets users and collects messages
• AI agent with a clear system message to guide menu, order, and status steps
• OpenAI chat model for natural language replies
• Window buffer memory that keeps recent context for smoother conversations
• HTTP tool to fetch the product menu from your endpoint
• HTTP tool to submit orders with the chat message mapped into the request body
• HTTP tool to check order status and return updates to the user
• Calculator tool for quick price and quantity math inside the chat

• Reduce manual order intake from 5 minutes to 1 minute per customer
• Automate up to 90 percent of repeat questions about menu and order status
• Improve order accuracy by 80 percent through structured prompts and API checks
• Handle 10 times more concurrent chats without adding staff
• Connect chat, ordering, and status checks in one place for faster service

1. Import the template into n8n: Create a new workflow in n8n, click the three dots menu, select 'Import from File', then choose the downloaded JSON file
2. You'll need accounts with OpenAI. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, create OpenAI credentials. If unsure, open the Chat OpenAI node, choose the credential dropdown, click Create new credential, then follow the on screen steps to add your API key from the OpenAI website.
4. Open the When chat message received node. Keep Public set to true to allow a public chat link. Edit the greeting to match your brand.
5. Open the AI Agent node. Review the system message and update product rules, ordering steps, and tone. Keep the prompt type as define.
6. Open the Chat OpenAI node and select your OpenAI credentials. Choose the desired model and temperature for your use case.
7. Open Get Products. Replace the sample URL with your menu endpoint if needed. Test the URL in your browser or with a tool like curl to confirm it returns data.
8. Open Order Product. Confirm method is POST and that the body sends the chatInput field. Adjust your backend to parse this field and extract name, product, and quantity.
9. Open Get Order. Point the URL to your order status endpoint. Decide how users will reference their order, such as name or order ID, and align it with your backend.
10. Check the Window Buffer Memory node size so it holds enough recent messages. A small window keeps the chat focused and fast.
11. Save the workflow and click Execute to preview. Use the public chat link to run a few sample conversations.
12. If you see 401 errors from the model, recheck your OpenAI key. If HTTP tools fail, verify the endpoint URLs and that they return valid JSON.