# Chat AI Agent Configuration

#### Overview

After creating your Chat AI Agent, you can customize various settings to optimize performance, align with your brand voice, and ensure seamless customer experiences.

***

#### Accessing Agent Settings

1. Navigate to **AI Agents & Bots** in the navbar
2. Click on your Chat AI Agent

***

#### Configuration Options

**1.  Prompt**

The system prompt is the foundation of your AI agent's behavior. It defines:

* **Role**: Who the AI is (e.g., "You are a helpful sales assistant...")
* **Goal**: Whats the Goal of the AI  (e.g., "Your goal is to qualify leads...")
* **Instructions**: What the AI should or shouldn't do
* **Knowledge Scope**: What information the AI has access to

<figure><img src="/files/b7Yf6StDTh1MgRKxbEkd" alt=""><figcaption></figcaption></figure>

**Best Practices:**

* Keep the prompt clear and specific
* Include examples of desired responses
* Define boundaries (what not to answer)
* Update regularly based on customer interactions

***

**2. Knowledge Base Management**

Add supplementary information to enhance your AI's knowledge:

**Sources:**

* Website URLs (auto-scraped during setup)
* Uploaded documents (PDFs, Word docs)

**Managing Knowledge:**

* Add new information as products/services evolve
* Remove outdated content
* Organize by categories for better retrieval

<figure><img src="/files/w19LsVowuJDXIhBdKbkx" alt=""><figcaption></figcaption></figure>

***

**3. Agent settings**

| Setting                              | Description                                                                                                      |
| ------------------------------------ | ---------------------------------------------------------------------------------------------------------------- |
| **Session Timeout**                  | Time before conversation ends due to inactivity                                                                  |
| **AI model & Document Intelligence** | Allows user to choose from available LLM model and also you can customise Knowledge base chunking configuration. |
| **AI processing message**            | Message displayed when Ai takes more than 10 sec to respond (you can enable or disable it)                       |
| **AI Image Recognition**             | Allow AI to understand and respond to images? (Available in whatsapp alone)                                      |

<figure><img src="/files/QUOiUWm1PAQPA0wneNO9" alt=""><figcaption></figcaption></figure>

***

**4. Human Handoff**

Configure when and how conversations transfer to human agents: You can use **@assign\_to** system action and select particular team member to escalte conversation to Human

**Human Handoff Triggers:**

* Customer requests to speak with a human
* Complex queries beyond AI scope

**Handoff Message:**

> "I'm connecting you with one of our team members who can better assist you. Please hold on..."

<figure><img src="/files/r5VkyZjzVfzjaUDyC3eU" alt=""><figcaption></figcaption></figure>

***

**5. Fallback Configuration**

Define what happens when the AI cannot answer a query:

**Options:**

* **Show Fallback Message**: Display a predefined message
* **Transfer to Human**: Route to a live agent
* **Ask for Clarification**: Request more details from the customer

**Recommended Fallback Message:**

> "I'm not sure I understand. Could you rephrase that, or would you like me to connect you with a team member who can help?"

***

#### Best Practices

1. **Start Simple**: Begin with default settings and refine based on real interactions
2. **Monitor Regularly**: Review conversations weekly to identify improvement areas
3. **Update Knowledge**: Keep business information current
4. **Train Your Team**: Ensure human agents understand AI handoff processes
5. **Gather Feedback**: Use customer feedback to improve prompts and responses


---

# 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.gallabox.com/chat-ai-agent/chat-ai-agent-configuration.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.