One of the most common automation patterns is handling inbound phone calls and routing the results to a messenger your team already uses. In this use case, you will set up an AI agent that answers incoming calls, follows a structured conversation plan, and sends a summary to a Telegram chat once the call ends.
This is useful for teams that want to automate first-line phone support, lead qualification, or appointment scheduling — without building a custom call center.
A fully automated inbound call pipeline:
Once set up, every incoming call is handled by the agent autonomously. Your team receives a Telegram message with the call outcome — no manual work required.
Incoming call → Phone number → AI Agent → Conversation → Call ends → Telegram notification
Step by step:
Go to the Agents section and create a new agent. Give it a clear name that reflects its purpose — for example, "Inbound Support Agent" or "Lead Qualifier".
Write an instruction for the agent that describes:
Keep the instruction concise and goal-oriented. The agent will follow it as a guide, not a rigid script.
Go to Integrations and create a new Telegram integration. You will need to connect it to a Telegram bot and specify the chat or group where notifications should be sent.
The integration defines what information is included in the notification — for example, caller intent, collected data, or call outcome.
Attach the Telegram integration to your agent. This tells the system to trigger a Telegram notification whenever this agent finishes a call.
Go to Phone Numbers and create or assign a number to your agent. Once linked, any call to this number will be handled by the agent automatically.
Call the number from your phone. Verify that:
When an incoming call arrives:
The entire pipeline runs without human intervention. The only manual step is the initial setup.
This pattern scales naturally. Whether you receive 10 calls or 10,000, the system handles them the same way.