# Use RCS with an AI Agent

One of the most efficient ways to use RCS is to set up an AI Agent to handle conversations with your customers.

AI Agents can reply to messages, send information about products and services, share payment links, upsell, cross-sell, and perform a variety of tasks.

## How to set up an AI Agent for RCS

In this guide, we'll give an overview of the steps you'll need to take to set up an AI Agent on RCS.

{% hint style="info" %}
For in-depth information on knowledge bases, agent configuration, and agent deployment, see the [AI Agents](/applications/ai-features/ai-agents.md) section of the docs.
{% endhint %}

#### What you'll need

* An OpenAI account
* An [RCS Agent](/applications/channels/channels/supported-channels/google-rcs/create-a-google-rcs-agent-and-install-google-rcs-as-a-channel.md), and [RCS installed as a channel](/applications/channels/channels/supported-channels/google-rcs/create-a-google-rcs-agent-and-install-google-rcs-as-a-channel.md#step-4-install-rcs-as-a-channel)

### Step 1: Install OpenAI integration

The first step you'll need to take is to install the [OpenAI](/applications/integrations/integrations/supported-integrations/openai.md) integration. This is the tool that will power your AI Agent.

<table data-view="cards"><thead><tr><th></th><th data-hidden></th><th data-hidden></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>Open AI →</strong> </td><td></td><td></td><td><a href="/pages/WwCL2FtbahXLiEijUS47">/pages/WwCL2FtbahXLiEijUS47</a></td></tr></tbody></table>

### Step 2: Create a knowledge base

Next, you'll need to [create a knowledge base](/applications/ai-features/ai-agents/how-to/set-up-your-ai-employees-knowledge-base.md) of information that your AI Agent will use to talk with your customers.&#x20;

Knowledge bases are essentially a collection of organized folders containing individual articles, much like a help center.

By building a well-organized and comprehensive knowledge base, your AI Agent will find it easy to access and distribute information.

<table data-view="cards"><thead><tr><th></th><th data-hidden></th><th data-hidden></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>Concept: Knowledge bases →</strong></td><td></td><td></td><td><a href="/pages/PYRCEON6q5M7FvaQ1nKO">/pages/PYRCEON6q5M7FvaQ1nKO</a></td></tr><tr><td><strong>How to: Build a knowledge base →</strong></td><td></td><td></td><td><a href="/pages/sV7qGC5ILS7fzXfaOLz4">/pages/sV7qGC5ILS7fzXfaOLz4</a></td></tr></tbody></table>

### Step 3: Set up an AI Agent

Now that you've added all of the information to a knowledge base, you'll need to [set up your AI Agent](#step-3-set-up-an-ai-agent).

In this step, you'll establish the AI Agent's personality and tone, what it is and is not allowed to do, its purpose, handover to a human protocol, and a variety of other functional properties.

{% hint style="info" %}
RCS Agents and an AI Agent are different concepts. [RCS Agents](https://developers.google.com/business-communications/rcs-business-messaging/guides/build/agents?model=xion) are a Google concept, whereas [AI Agents ](/applications/ai-features/ai-agents.md)are a Bird concept.
{% endhint %}

<table data-view="cards"><thead><tr><th></th><th data-hidden></th><th data-hidden></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>How to: Set up an AI Agent →</strong></td><td></td><td></td><td><a href="/pages/57Ub98FzBoKPzzKQjOiG">/pages/57Ub98FzBoKPzzKQjOiG</a></td></tr></tbody></table>

### Step 4: Deploy your AI Agent

The deployment process for AI Agents is the same, regardless of the channel you want to deploy it on.

You need to [build a flow](/applications/ai-features/ai-agents/how-to/deploy-an-ai-agent.md), and set the trigger so that the flow runs when a new message is received on your installed RCS channel.

Once published, your AI Agent will automatically reply to messages that you receive on your RCS channel. You'll be able to see these conversations, and take over manually as required, in Inbox.

<table data-view="cards"><thead><tr><th></th><th data-hidden></th><th data-hidden></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>How to: Deploy an AI Agent →</strong></td><td></td><td></td><td><a href="/pages/C0yXxP7G7b97s3kViVHW">/pages/C0yXxP7G7b97s3kViVHW</a></td></tr></tbody></table>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.bird.com/applications/channels/channels/supported-channels/google-rcs/use-rcs-with-an-ai-agent.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.