Chat Trigger

The Chat Trigger gives your automation a hosted chat page or embedded endpoint people can talk to. Each accepted message starts a run, and the final Step’s output supplies the reply.

"Can you check the status of order #4821?"

Setting it up

What it passes on

The last Step’s output is sent back as the reply, so the caller waits for the run to finish. Dropping {{ N.historyText }} into an AI Step’s instructions lets the model see what was already said in the same session.

Giving it a memory

Conversation memory (messages) sets how many earlier messages of the same sessionId are handed to the automation as history. Set it to 0 to omit earlier chat messages from the trigger’s history. An AI Agent’s separately configured memory can still supply context.

Longer history adds context and increases prompt size. Choose a window that covers the follow-up questions your automation needs to handle.

Keeping the cost predictable

Every message runs the automation and spends your team’s credits, so an open link is a running cost as well as an open door.

Messages per visitor per minute caps how fast one person can send. A per-visitor rate limit is always applied when Authentication is None, and this setting is how you tighten it. For an internal tool, Glow User Auth is the stronger control: only your team can start a run at all.

Making the page your own

Four settings shape the hosted page:

Examples to copy

A support bot over your own documentation

Chat Trigger → AI Prompt. In the instructions:

Answer using only the product documentation supplied below.
If the answer is not present, say what information is missing.

Product documentation: [insert approved documentation text here]

Conversation so far:
{{ 1.historyText }}

Question: {{ 1.chatInput }}

Set Conversation memory to 10 so follow-up questions like “and what about the Pro plan?” still make sense.

An internal assistant, restricted to your team

Set Authentication to Glow User Auth and share the link internally. Only authorized workspace members can use it. Authentication does not remove rate limits or usage costs; choose limits appropriate to the workload.

A chat window inside your own product

Set Mode to Embedded Chat and POST each message as JSON from your widget:

{
  "chatInput": "Where is my latest invoice?",
  "sessionId": "sess_9f8a7b6c5d4e",
  "metadata": { "customerId": "cus_42", "page": "/billing" }
}

chatInput is the only required field. Send the same unguessable sessionId for each message from the same visitor so their conversation history continues (omit it to start fresh, and never reuse an ID across different visitors). Read custom metadata fields in the automation as {{ 1.metadata }}.

What’s Next?

• AI Prompt: the step that usually writes the reply.
• Webhook Triggers: for starting an automation from a system rather than a person.
• Variable Syntax: the full reference for {{ }}.