Skip to main content
Automations let your AI agent do more than just reply. When a customer starts a conversation, asks about a product, or matches a keyword, ChatRos can automatically add them to a Mailchimp list, send an email, call a webhook, or even reply with a scripted message — all without human intervention.

​
How Automations Work

Every automation has two parts: a trigger (what starts it) and one or more actions (what happens next). You connect them visually in a node-graph editor by drawing lines between nodes. Automations run at two points in every conversation:
  • Before the AI responds — If your automation contains a Reply action, ChatRos sends that scripted reply and skips the AI entirely. This is useful for greeting messages, lead capture forms, or FAQ shortcuts.
  • After the AI responds — All other actions (Mailchimp, MailerLite, email, webhook) run automatically after the AI has replied.

​
Trigger Types

Conversation Started

Fires once when a customer sends their very first message. Ideal for welcome replies and lead capture.

Product Inquiry

Fires when the AI searches for products or the customer’s message contains words like “buy”, “price”, or “item”.

Order Inquiry

Fires when the AI looks up an order or the customer mentions “order”, “tracking”, or “delivery”.

Custom Keyword

Fires when the customer’s message contains keywords you define. Supports any (one match) or all (every keyword must match) modes.

​
What You Can Do with Actions

Send a scripted text message, share a file, or display a dynamic form that collects information like name, email, or order number.
Automatically subscribe customers to a Mailchimp audience or MailerLite group — and apply or remove tags — when they engage with your agent.
Fire an internal email alert to your team whenever a high-value conversation event occurs.
Send conversation data to any external system via HTTP GET, POST, PUT, or DELETE. Supports bearer token, API key, and basic auth.
Add an If/Else node to route the workflow differently depending on customer data — for example, handle first-time vs. returning customers differently.
Insert a custom AI prompt mid-workflow to generate a summary, classify intent, or produce dynamic content that feeds into later nodes.

​
Workflow Types

ChatRos supports two workflow modes. All new automations use the node-graph editor.
ModeDescription
Node GraphVisual canvas with drag-and-drop nodes and edges. Supports conditions, AI steps, and multi-action flows.
LegacySimple single-trigger → single-action format. Still supported for existing rules.
New automations always use the node-graph editor. Legacy rules created before the editor was introduced continue to work unchanged.

​
What’s Next?

Building Workflows

Learn how to create a node-graph automation step by step

Template Gallery

Start from a ready-made template instead of building from scratch

Mailchimp Integration

Connect your Mailchimp account to use in automations

MailerLite Integration

Connect your MailerLite account to use in automations