# Support Agent

- **URL:** https://agentline.cloud/build/support-agent
- **Category:** Support
- **Keywords:** AI inbound customer support, AI phone support agent, 24/7 AI call center, AI voice API, phone number for AI agents

> Triage, respond, escalate, follow up, close

A 24/7 inbound support line. The agent authenticates the caller, answers routine questions, and escalates anything it cannot resolve — with the transcript attached.

## How it works

1. Buy a number and point callers (or a support IVR) at it.
2. The agent greets, identifies the issue, and answers from your prompt.
3. Routine questions close on the call. Ambiguous ones escalate.
4. Every call writes a transcript you can ticket or review.

## Greeting

Thanks for calling support. What can I help with?

## System prompt

```
You are an inbound customer support agent on a live call. Keep every spoken turn under 15 words.

You can handle: order status, password reset steps, billing due dates, business hours, and how to reach a human.

Rules:
- Ask for an order ID or account email when needed.
- Never invent order status, refunds, or account balances.
- If you do not know, say so and offer a human.
- Escalate immediately for legal, medical, or payment-dispute issues.
- End by confirming the caller is done.
```

## Python

```python
from agentline import AgentLine

client = AgentLine(api_key="al_live_...")

agent = client.agents.create(
    name="Support Agent",
    voice_id="female-1",
    initial_greeting="Thanks for calling support. What can I help with?",
    system_prompt="""You are an inbound customer support agent. Keep turns under 15 words. Handle order status, password reset steps, billing due dates, and hours. Never invent account data. Escalate legal, medical, or payment disputes.""",
)

number = client.numbers.buy(agent_id=agent.id, country="US", area_code="415")
print("Number:", number)
```

## Node.js

```ts
import AgentLine from "agentline";

const client = new AgentLine({ apiKey: "al_live_..." });

const agent = await client.agents.create({
  name: "Support Agent",
  voiceId: "female-1",
  initialGreeting: "Thanks for calling support. What can I help with?",
  systemPrompt: `You are an inbound customer support agent. Keep turns under 15 words. Handle order status, password reset steps, billing due dates, and hours. Never invent account data.`,
});

const number = await client.numbers.buy({
  agentId: agent.id,
  country: "US",
  areaCode: "415",
});
console.log("Number:", number);
```

## Tell your AI

```
Set up a 24/7 inbound support phone agent on AgentLine. Follow https://agentline.cloud/skill.md then apply https://agentline.cloud/build/support-agent.md
```

## FAQ

**Can an AI agent answer inbound support calls?**

Yes. Provision a number, set this prompt, and AgentLine transcribes the caller in real time. Your agent replies with text; we speak it back.

**How does escalation work?**

Tell the agent exactly which topics need a human. It collects a callback number, summarizes the call, and ends. You pick the summary up from the transcript or a webhook.

**Is this a replacement for a call center?**

It handles the volume so humans handle exceptions. Teams typically deflect 60–70% of routine calls and keep a person for the rest.