Your First Pipeline

Create a working pipeline in 5 minutes: when a new deal appears in AmoCRM, send a notification to Telegram.

Prerequisites

• AmoCRM and Telegram Bot connected to Triggo

Step 1 — Describe Your Task

Open the chat and describe your task in plain language. For example:

When a new deal is created in AmoCRM, send a Telegram notification with the deal name and amount

No special syntax needed — AI understands natural language. Describe the task as you would explain it to a colleague.

Step 2 — AI Creates the Pipeline

Triggo will analyze your request and generate a pipeline. A pipeline card will appear in the chat:

• Trigger: AmoCRM → new deal
• Action: Telegram Bot → send message

The pipeline uses data from the trigger: deal name and amount. The Telegram message template is generated automatically — for example: "New deal: {name}, amount: {amount}".

Step 3 — Connect Services

If AmoCRM and Telegram are not yet connected, AI will prompt you to authorize right in the chat. Follow the connection instructions.

If the services are already connected, this step is handled automatically.

Step 4 — Activate

Click "Activate" on the pipeline card. The status will change to "Active" — the pipeline starts listening for events from AmoCRM.

An active pipeline runs in the background: every time a new deal is created in AmoCRM, Triggo automatically sends a notification to Telegram.

Step 5 — Test It

Create a test deal in AmoCRM. Within a few seconds, the Telegram bot will send a message with the deal data.

Under the hood:

1. AmoCRM notifies Triggo about the new deal (via webhook)
2. Triggo extracts the name and amount from the deal data
3. The bot sends the formatted message to Telegram

What's Next

• Modify the pipeline — tell AI what you want to change. For example: "Add the responsible manager's name to the message" or "Only send deals with amount over 100,000"
• View execution history — ask AI "show run history" to see all pipeline executions
• Something went wrong? — check the troubleshooting guide or ask AI "why did it fail?"