# n8n Chat on Your Website: Make Chat Publicly Available

> Put an n8n AI agent chat on your website: turn on Make Chat Publicly Available, publish, link or embed it, and test it like a customer first.

Source: https://learn.ynteractive.com/content/blog/n8n-public-chat-website · Updated 2026-10-20 · Free from Stepthrough (https://learn.ynteractive.com)

Quick answer

To put an n8n chat agent on your website, open the **When chat message received** trigger, turn on **Make Chat Publicly Available**, then **Publish** the workflow. n8n gives the chat a public URL you can link to, and the trigger's settings show how to embed it on a page. Test it with **Open chat** first, like a customer would. The free [Support Agent That Answers from Your Own Docs](https://learn.ynteractive.com/content/support-agent-answers-from-your-docs-n8n) lesson ends with exactly these steps.

Last updated October 20, 2026

[Take the free module →](https://learn.ynteractive.com/courses/ai-agents-101/9)Module 9 of the free [AI Agents 101](https://learn.ynteractive.com/courses/ai-agents-101) course, about 13 minutes. No account needed.

By [Robert Breen](https://learn.ynteractive.com/content/about) · Published October 20, 2026 · Updated October 20, 2026 · 3 min read

n8n · Chat · AI agents

## What does "Make Chat Publicly Available" do?

It lets people who aren't logged in to n8n use the chat. Without it, the chat only works inside the editor for testing. The [chat trigger](https://learn.ynteractive.com/content/glossary/chat-trigger) starts the workflow once per message.

## How do I publish the chat?

1. Double-click **When chat message received**.
2. Turn on **Make Chat Publicly Available**.
3. Close the node and click **Publish**.
4. Keep the **Version name** and click **Publish** again.

Until the workflow is published, the public chat doesn't answer. n8n [Publish Workflow](https://learn.ynteractive.com/content/glossary/publish-workflow) explains what publishing changes.

## Hosted chat or embedded chat?

| Option | What visitors get | Good for |
| --- | --- | --- |
| Hosted chat | A full-page chat at n8n's public URL | A "Chat with us" link in emails or on a contact page |
| Embedded chat | A chat bubble on your own site | A help widget on every page |

The trigger's mode setting switches between the hosted page and embedding, and n8n's chat package provides the snippet for your site.

## How should I test before going live?

Click **Open chat** in the editor and play the customer. In the lesson you test the property manager's agent as a tenant: questions the FAQ answers, and one that needs a person, which should trigger the Gmail handoff to the office. Then close the chat and check the execution: which tools did the agent use?

A good test list:

- Three questions the doc answers.
- One question it doesn't, which should get "I don't know" or a handoff.
- One emergency, which should get the emergency instruction first.
- One attempt to get it off topic.

## Try it free, step by step

**A Support Agent That Answers from Your Own Docs**. Turn an n8n chat agent into a support agent that answers only from your policies in a Google Doc, refuses what it should, and hands off to a person by email. You do every step yourself in a practice copy of n8n, it checks your work as you go, and “Do it for me” finishes any step you get stuck on. Free, and nothing touches your real accounts.

[Start the free lesson](https://learn.ynteractive.com/modules/support-agent-docs)

[Or read the full step-by-step guide](https://learn.ynteractive.com/content/support-agent-answers-from-your-docs-n8n)

## What should I set before strangers use it?

- **Initial message.** Say it's an [AI assistant](https://learn.ynteractive.com/content/glossary/ai-assistant) and what it can help with.
- **Limits in the system message.** No legal advice, no promises, short answers.
- **Authentication if it's internal.** The trigger can require basic auth or n8n login for staff-only chats.
- **A handoff path** to a person, as in the lesson.
- **A look at costs.** Every message calls the model. Watch usage in the first weeks.

## Does the chat remember the conversation?

Only if the agent has memory. With **Simple Memory** connected, the [chat trigger](https://learn.ynteractive.com/content/glossary/chat-trigger)'s session ID keeps each visitor's conversation separate. [n8n Simple Memory](https://learn.ynteractive.com/content/blog/n8n-simple-memory) explains how.

## Key takeaways

- Make Chat Publicly Available plus Publish puts the chat on a public URL.
- Use hosted mode for a link, embedded mode for a widget on your site.
- Test with Open chat as a customer, including a question the doc can't answer.
- Add an initial message, limits and a handoff before going live.

## Frequently asked questions

**How do I make an n8n chat public?**

In the When chat message received trigger, turn on Make Chat Publicly Available and publish the workflow.

**Why doesn't my public n8n chat respond?**

The workflow isn't published, or the public setting is off. Publish it and try the public URL again.

**Can I restrict the chat to my team?**

Yes. The chat trigger supports authentication, so you can require a login instead of leaving it open to everyone.

**Which course is this from?**

[AI for Customer Support](https://learn.ynteractive.com/courses/ai-for-customer-support) and [AI Agents 101](https://learn.ynteractive.com/courses/ai-agents-101). The [live chat](https://learn.ynteractive.com/content/glossary/live-chat) glossary entry covers human chat.

## Free lessons in this post

- [A Support Agent That Answers from Your Own Docs](https://learn.ynteractive.com/content/support-agent-answers-from-your-docs-n8n): Turn an n8n chat agent into a support agent that answers only from your policies in a Google Doc, refuses what it should, and hands off to a person by email.

## Watch it in a free lesson: A Support Agent That Answers from Your Own Docs

### A Support Agent That Answers from Your Own Docs: video walkthrough

Full walkthrough · 2:23 · with captions

Turn an n8n chat agent into a support agent that answers only from your policies in a Google Doc, refuses what it should, and hands off to a person by email. The whole lesson narrated step by step, recorded in Stepthrough’s simulated practice environment.

[Watch it with chapters and the full transcript](https://learn.ynteractive.com/content/support-agent-answers-from-your-docs-n8n#watch)

From the free lesson [A Support Agent That Answers from Your Own Docs](https://learn.ynteractive.com/content/support-agent-answers-from-your-docs-n8n)

## Related terms

Plain-English definitions, each with an example from a free lesson.

[Publishing a workflow](https://learn.ynteractive.com/content/glossary/publish-workflow) · [AI assistant](https://learn.ynteractive.com/content/glossary/ai-assistant) · [Chat Trigger](https://learn.ynteractive.com/content/glossary/chat-trigger) · [Agentic RAG](https://learn.ynteractive.com/content/glossary/agentic-rag) · [Max iterations (agent step limit)](https://learn.ynteractive.com/content/glossary/max-iterations) · [Reranking](https://learn.ynteractive.com/content/glossary/reranking) · [Retrieval-augmented generation (RAG)](https://learn.ynteractive.com/content/glossary/rag) · [Support ticket](https://learn.ynteractive.com/content/glossary/support-ticket)

[The full AI glossary](https://learn.ynteractive.com/content/glossary)

## Keep reading

- [n8n Google Docs Tool: Let an AI Agent Answer from a Doc](https://learn.ynteractive.com/content/blog/n8n-google-docs-tool-agent): Give an n8n AI Agent your own FAQ with the Google Docs Tool (Operation: Get), a Gmail Tool handoff to staff and system message rules.
- [How to Automate Customer Support with AI (Safely)](https://learn.ynteractive.com/content/blog/how-to-automate-customer-support-with-ai): How to automate customer support with AI in four stages: sort incoming requests, draft replies, answer FAQs from your own docs, and hand off to a person.
- [How to Build an Internal AI Assistant for Your Team](https://learn.ynteractive.com/content/blog/how-to-build-an-internal-ai-assistant): How to choose a platform for an internal AI assistant: ChatGPT or Claude Projects, Gemini Gems or an n8n agent, plus five steps to build and test one.

[All blog posts](https://learn.ynteractive.com/content/blog) · [All free tutorials](https://learn.ynteractive.com/content)
