# Create AI Agent which replies to user greetings or thanks / good byes

### Step 1: Access the Playground

* Go to the "playground" section in your AI platform.​

### Step 2: Set System Instructions

* Instead of referring to any documents, select the option to write "system instructions".​
* Write detailed instructions such as:
  * Your agent’s name.
  * Desired language style (e.g., reply in soft language, reply precisely).​
  * Specify that the agent should reply to greetings with a greeting and a follow-up question.​
  * For example: "User will send you a greeting. Reply back with a greeting and a question about support, sales, etc.".​

### Step 3: Test Greeting Responses

* Send a greeting message (such as "hi") in the playground to see how your AI agent replies.​
* Modify the instructions as needed to adjust the tone and content of the greeting response.​

### Step 4: Create Conversation-End Responses

* Write instructions for ending conversations, such as replying to "thanks" or "bye".​
* For example, instruct the agent to say “You’re welcome,” or “Goodbye, see you later” when the user sends thank you or farewell messages.​

### Step 5: Experiment and Refine

* Try various messages and refine the instructions for both greeting and conversation-end scenarios to achieve the desired behavior.​
* You can play with the instruction details to customize for different use cases and personalities.​

{% embed url="<https://youtu.be/AWXFITI4gJE>" %}


---

# 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://helpdocumentation.gitbook.io/user/ai-dashboard/create-ai-agent-which-replies-to-user-greetings-or-thanks-good-byes.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.