# Agent Details

The **Agent Details** section is where you define the core configuration of your AI Agent.

This is the first step in creating an agent and determines:

* Which AI model powers the agent
* How the agent behaves
* What its purpose and identity are

Without completing this step, the agent cannot function.

***

### Overview

Agent Details consists of two main parts:

1. **Model Configuration** — defines the AI provider and model
2. **Agent Configuration** — defines the agent’s identity and behavior

***

### Model Configuration

This section defines which AI model your agent will use.

#### Provider

Select the AI provider:

* Azure OpenAI
* OpenAI

***

#### Credential

Choose a credential to authenticate with the selected provider.

Credentials store:

* API keys
* Endpoint configurations
* Access settings

> The agent will use this credential to communicate with the AI provider.

***

#### Model

Select the model to power your agent.

Examples:

* GPT-based models
* Provider-specific deployments (for Azure OpenAI)

> The selected model directly impacts response quality, speed, and cost.

***

### Agent Configuration

This section defines how your AI Agent behaves.

***

#### Agent Name

Provide a unique name for your agent.

This name is used to identify the agent across the platform.

***

#### Instructions (Role Prompt)

Define the behavior and role of the AI Agent.

You have two options:

**Use Ready-Made Instructions**

Select from predefined templates to quickly configure common agent types.

***

**Create Custom Instructions**

Write your own instructions to fully control:

* Agent personality
* Scope of knowledge
* Allowed actions
* Response style

***

> Instructions are the **brain of the agent**.\
> They determine how the agent interprets and responds to user input.

***

### Save Configuration

After completing both sections, click **Save** at the top of the page.

This finalizes the initial setup of your AI Agent.

***

### What’s Next

Once the Agent Details are configured, you can continue building your agent by:

* Adding [**Knowledge**](https://docs.autommate.com/userguide/ai-agent-composer/knowledge) (for context and RAG)
* Defining [**Skills**](https://docs.autommate.com/userguide/ai-agent-composer/skills) (for actions and integrations)
* Connecting [**Channels**](https://docs.autommate.com/userguide/ai-agent-composer/channels) (WhatsApp, Teams, Webhook, Mate Chat.)

***

### Summary

Agent Details is the **foundation layer** of your AI Agent.

* Model → defines intelligence
* Instructions → define behavior

Everything else (skills, channels, knowledge) builds on top of this.