> ## Documentation Index
> Fetch the complete documentation index at: https://docs.chatling.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Getting started with AI Agents
> Learn how to create your first AI agent in Chatling
What we'll be covering:
1. [Create an AI Agent](#1-create-an-ai-agent)
2. [Configure the AI settings](#2-configure-the-ai-settings)
3. [Populate the knowledge base](#3-populate-the-knowledge-base)
4. [Create AI actions](#4-create-ai-actions)
5. [Test your Agent](#5-test-your-agent)
6. [Deploy](#6-deploy-your-agent)
## 1. Create an AI Agent
1. Go to your account and open the "My agents" page. Click the `+ Create` or `New` button.
2. Select `AI Agent` as the type.
3. Enter a name for your Agent and click `Create agent`.
## 2. Configure the AI settings
Once your Agent is created, it's time to configure it's AI settings.
1. From your agent's dashboard, click `Settings` in the sidebar.
2. Select the `AI` menu.
3. Here you can configure the AI settings, such as:
* **AI Model**: The AI model that the Agent will use to generate responses.
* **Temperature**: Controls randomness of the agent's responses. A lower temperature will make the outputs more focused and deterministic, whereas a higher temperature will make the responses diverse and creative.
* **Instructions**: Define the agent's role, goals, guardrails, and style (often called the "system prompt"). There are templates available to help you get started. Click the `Browse templates` button to see the available templates.
## 3. Populate the knowledge base
The Knowledge Base is where you upload information the Agent uses to generate responses to user queries. You can upload information about your business, products, services, policies, and more.
You can add several types of data to the knowledge base, such as websites, documents, texts, and FAQs. You can also connect your Zendesk or Zoho account to import articles from your help center.
When a user asks a question, the Agent queries the knowledge base for relevant information and generates a response based on the data it finds.
To add data to the knowledge base, click the `+ Add data source` button, then select the type of data you want to add.
## 4. Create AI actions
Actions unlock the true power of AI Agents. Instead of only generating replies, the Agent can actively carry out tasks during a conversation.
This might involve saving details of potential leads, displaying booking widgets, interacting with external systems to fetch or store data, creating support tickets, and more.
1. Click the `Actions` menu from the sidebar.
2. Click `Create action`.
3. Select the type of action you want to create.
4. Configure the action according to your needs.
5. Click the `Create` button to save the action.
To learn more about actions, check out the [Actions](/ai-agent/actions/introduction) documentation.
## 5. Test your Agent
You can test your AI Agent in the Playground. It is your sandbox to experiment with your Agent in real time.
To open the Playground, click `Playground` from the sidebar menu.
You can tweak core settings—like AI model, enabled actions, temperature, and instructions—then chat with the Agent to see exactly how those changes affect behavior.
You can also run side-by-side comparisons of up to 5 instances of your Agent, each with different settings, to quickly identify the configuration that performs best.
## 6. Deploy your Agent
Once you've built your AI Agent, it's time to deploy it to your website, WhatsApp, or other channels.
Click `Deploy` from the sidebar menu. Select the channel you want to deploy your Agent to and follow the on-screen instructions.
