Resources · n8n integration

n8n Chat nodeConfigure Chat in n8n.

The n8n Chat node puts a human back in the middle of a running workflow. It sends a message into the chat, then either carries on or holds the execution until someone answers. Six parameters cover the text, the reply mode, the memory input and how long the wait lasts.

Try the Chat node on n8nSee the 6 parameters

Verified Trustpilot reviews · AI, automation & growth agency

Why automate

What does the n8n Chat node do in a workflow?

The Chat node sends a message into the chat and, when you ask it to, pauses the execution there until a person replies. It runs like any other action node: one message per incoming item, no sub-node ports of its own. Everything it can do sits in six parameters, from the text itself to the shape of the answer you expect back.

Automation stops being comfortable the moment a step needs judgment. Refunding a customer, sending a contract, deleting a row: the workflow knows how, it just should not decide alone. That is the gap this node fills, and it is why n8n files it under human in the loop.

First scenario, approval before an irreversible action. The node posts the summary through Message, Response Type is set to approval, and the run only continues once someone clicks. Pair it with a Slack alert earlier in the workflow and the reviewer knows something is waiting.

Second scenario, collecting the missing piece. Set Response Type to freeTextChat and the person types the answer straight into the chat, or fill in Form Elements so they pick from proper fields instead of guessing the format. Useful when a draft reply to a customer has to be approved before Gmail sends it.

Third scenario, a conversation rather than a single question. Turn on Add Memory Input Connection and the node gets a memory input, so a sub-node (a small node that plugs into a port of another node rather than running on its own) can keep the thread of what was already said.

When to use something else: if nothing needs a human answer, a plain notification is lighter, and an API that n8n has no dedicated node for belongs in HTTP Request. Known limits: the node describes version 1, it has no sub-node port beyond the optional memory input, and a run that waits still counts as a run that has not finished. Budget for that before you wire it into something that fires all day, or work through it with the team during the n8n training.

Parameters

What does each Chat parameter control?

The Chat node has 6 parameters. For each one: the node as you configure it in n8n, what the parameter changes, and our field notes.

01

Message

message

What you see in n8n

Notes & use cases

Holds the text the node posts into the chat. It is the only required parameter, so an empty value stops the node from running.

Key parameters

  • Message: a string, usually built from the item that arrived, for example {{ $json.summary }} so the reader sees the record rather than a generic prompt.
Use cases
a refund request quoting the amount and the customer name, so whoever answers has everything on screen and does not have to open another tab.
02

Wait for User Reply

waitUserReply

What you see in n8n

Notes & use cases

Decides whether the execution stops at this node. Left off, the message goes out and the workflow keeps moving; switched on, the run pauses until an answer comes back.

Key parameters

  • Wait for User Reply: a boolean, the switch between a one way message and a real checkpoint.
Use cases
a nightly job that posts its result and moves on stays off; a contract about to be sent needs it on, because nothing downstream should happen without a yes.
03

Response Type

responseType

What you see in n8n

Notes & use cases

Sets the shape of the answer expected from the person. Two modes exist, and each one opens its own set of fields.

Key parameters

  • Response Type: approval lets the reader approve or disapprove from inside the message, freeTextChat lets them submit a written response in the chat.
  • Block User Input: blocks input from the user while an approval is pending.
  • Define Form: fields to build the form from the elements below, json to describe it as JSON.
  • Form Fields, Form Elements, Approval Options: the JSON definition, the form elements themselves, and the approval choices offered.
Use cases
approval for a go or no go, free text when the answer is a reason that has to be written down.
04

Add Memory Input Connection

options.memoryConnection

What you see in n8n

Notes & use cases

Adds a memory input to the node, so a memory sub-node can hold what was already exchanged instead of treating every message as the first one.

Key parameters

  • Add Memory Input Connection: a boolean; off, the node posts and forgets, on, a port appears for the memory sub-node.
Use cases
a back and forth where the reader answers with a question of their own, and the next message has to make sense of what came before rather than starting over.
05

Limit Wait Time

options.limitWaitTime

What you see in n8n

Notes & use cases

Caps how long the node waits for a user response before the execution resumes. Without it, a run can sit there for as long as nobody answers.

Key parameters

  • Limit Type: afterTimeInterval waits for a certain amount of time, atSpecifiedTime waits until a set date and time.
  • Amount and Unit: the time to wait, counted in minutes, hours or days.
  • Max Date and Time: continues the execution after that date and time.
Use cases
a question sent on a Friday afternoon that should not block the workflow all weekend.
06

Auto-save highlighted data

options.autoSaveHighlightedData

What you see in n8n

Notes & use cases

Saves highlighted data automatically, which is what lets you filter executions later in the Executions view. It defaults to true.

Key parameters

  • Auto-save highlighted data: a boolean, available on Pro and Enterprise plans in n8n Cloud, and on Enterprise or registered Community Edition when self-hosted.
Use cases
finding, weeks later, which approvals went through and which ones timed out, without scrolling the whole execution list one by one.
Need help

Need help automating Chat with n8n?

A person reads every message.

FAQ

Chat node questions

01Is the Chat node included in n8n?
Yes. It belongs to the n8n AI package, shipped with n8n, so there is nothing to install and nothing extra to pay on the n8n side. It works the same on n8n Cloud, the hosted offer run by n8n, and on a self-hosted instance under the Community Edition and its Sustainable Use license. One option is narrower: auto-saving highlighted data is available on Pro and Enterprise plans in n8n Cloud, and on Enterprise or registered Community Edition when you host it yourself. If a model provider is involved elsewhere in your workflow, that provider bills its own API calls under its own terms.
02What credentials does the n8n Chat node need?
None. There is nothing to authenticate on this node: it has no credential and no Authentication selector, which is why the page has no connection section at all. The account side of things lives with the model sub-node plugged into the rest of your workflow, and that sub-node carries the provider credential. So setting this node up is only about filling parameters: a Message, a Response Type if you want a structured answer, and a wait limit if you do not want a run hanging around. Nothing to create, nothing to paste, nothing to rotate.
03What are the limits of the Chat node?
This page describes version 1 of the node, and an older workflow may show an earlier version with fewer options. The node runs like a regular action node, once per incoming item, and it has no sub-node ports of its own beyond the memory input you switch on with Add Memory Input Connection. It does not decide anything by itself either: it asks, and the branch you build after it handles the answer. And a paused execution is an open execution, so set Limit Wait Time whenever nobody is guaranteed to reply.
04What do you need to plug in to make it work?
At a minimum, a Message. That is the only required parameter, and the node will not run without it. From there you add what the situation calls for: Wait for User Reply to actually pause the run, Response Type set to approval for a yes or no, or to free text when you need a written answer. To keep a conversation coherent across several messages, switch on Add Memory Input Connection and connect a memory sub-node to the port that appears. Start with message plus approval, that combination covers most human checkpoints.
05n8n or Make for a human approval step?
It depends on four things. Hosting: n8n runs on your own server through Docker or npm, or on n8n Cloud, while Make is hosted only. Data control: a self-hosted instance keeps the conversation and everything around it on your infrastructure. Cost model: Make bills per operation, which adds up when a workflow loops. Visual logic: both are canvas based, and the answer usually comes down to how comfortable your team is reading a branch. Pick Make if hosting is not your problem, and n8n if you want the run and its data under your own roof.
Hack'celeration Lab

Get our weekly integration tips.

No spam. Unsubscribe anytime.

Same platform

Other docs you might need on this stack.

Browse all integrations