Package Information
Documentation
n8n-nodes-msg-buffer
This is an n8n community node. It lets you use a timed message buffer in your n8n workflows.
The timed message buffer collects incoming messages over a sliding time window — the window resets with each new message — and emits them all at once when no new messages arrive within the configured interval.
Installation
Operations
Credentials
Compatibility
Usage
Resources
Installation
Follow the installation guide in the n8n community nodes documentation.
Operations
- Buffer messages — Accumulate incoming data under a shared session key. Each new message resets the time window.
- Emit buffered messages — Once no new messages arrive within the configured interval, release all collected messages together through the Ready output.
Credentials
This node requires a Redis credential to store buffer state across executions.
Without a valid Redis credential, the node will not function.
Compatibility
- Last tested on: n8n
1.100.0 - Minimum version: Not determined — use
1.100.0or later to be safe.
Usage
The node has five parameters:
| Parameter | Type | Description |
|---|---|---|
| Session Key | String | Unique identifier to group messages into the same buffer (e.g. conversation-123). |
| Content | String | The data you want to accumulate (e.g. the message text). |
| Wait Amount | Number | How long to wait after the last message before emitting (e.g. 5). |
| Wait Unit | Dropdown | Unit for the delay: seconds, minutes, hours, or days. |
| Redis Credential | Credential | The Redis connection to use for storing buffer state. |
Outputs
- Ready — Fires once after the wait interval expires with no new incoming messages. Returns an object with all buffered content:
{
"data": [
"Hi!",
"How are you?",
"What's up?"
]
}
- Discarded — Fires on every execution that arrives while a window is already running. Returns an empty object
{}and does nothing else.
When should you use this node?
This node is useful whenever you receive a burst of related messages in a short period and want to process them together instead of one by one.
Example: Chat Message Buffering

With n8n-nodes-msg-buffer:
- The first message opens a 5-second window and is added to the buffer.
- The next four messages arrive during that window, each resetting it, and are routed to Discarded — they are buffered silently, no action is taken yet.
- After 5 seconds of inactivity from the last message, the node fires the Ready output with all five messages:
{
"data": [
"Hi!",
"How are you?",
"What's up?",
"Let's meet.",
"Bye!"
]
}
Your workflow then processes them as a single, complete message — one AI call, one coherent reply.