You can shape the AI agents interaction behavior and improve consistency by following some of the best practices when you write guide instructions. The predefined commands allow you to structure your guide in such a way that your AI agent interactions are reliable and consistent throughout the interaction sessions.

Plan your guides

As you plan your guide, make sure to stay on goal and ensure that your guides are not lengthy.

Keep the guide focused on a single task

Genesys recommends that you design each guide to handle a single, well-defined customer intent such as “change delivery date” or “check account balance”. Limiting each guide with one intent narrows the scope of the large language model (LLM) to ensure predictability and accuracy. For conversations involving multiple intents, connect multiple guide blocks within a single Architect flow.

Keep the guide short

To build a successful guide, start with the essential steps first, validate with real users and data, and slowly extend the capabilities to achieve the preferred result. With the right foundations and an iterative approach, AI agents can deliver real business value, automating not just tasks, but entire workflows with intelligence and adaptability. 

Maintain a professional and conversational tone

While your guide’s instructions can be technical, the resulting AI agent interaction should be natural. Phrase your instructions in such a way that they create a professional, helpful, and conversational tone.

Build your guides

Follow these guidelines for structuring, formatting, and creating simple, manageable tasks.

Structuring and formatting your guide 

Use the predefined commands as building blocks for your guide as you write instructions of your guide.

Break down your task into simple steps

Divide a complex task into simpler, more manageable steps. This process ensures that your Virtual Agent can follow the instructions accurately without getting overwhelmed.

• Name each step in the order you want your AI agent to run the task. Naming the steps also allows you to instruct your AI agent to break sequential execution and jump to any of the steps in the guide.
• As an added measure, segregate your guide into main steps and substeps. With this approach, you can assign a unique name to your main steps, allowing your AI agent to go to any of the main steps as necessary.

Introduction
- Say "Hello, I'm ACME's Router Connectivity assistant. What can I help you with?"

Basic Checks
- Ask "Let's start with some basic checks. First, please verify that your router is turned on and all cable connections are secure."
- Store in {{Variable.router_on}} and {{Variable.cable_check}} 
- If cable connections are not secured
- Then Go To Step Plug Cables
- Else Go To Step Ask Power Status Lights

Plug Cables
- Ask "Please secure the cables. Has that resolved the issue?"
- Store in {{Variable.cable_fixed}} 
- If cable connections are secured and it resolved the router issue
- Then Go To Step Resolution
- Else Go To Step Ask Power Status Lights

Ask Power Status Lights
...
        

Use Go to command to jump between steps

• Refer to other steps using step names. 
• State the destination step name clearly.

Reduce ambiguity

• When defining steps, be precise and specific about the outcome or response you want from the AI agent.
• Abstain from using ambiguous, vague, or broad instructions. Clear instructions help your AI agents to produce better results.
• Provide as much context as possible by including variables, data actions, and specific guidance on handling different scenarios. Set up your agents to handle unexpected scenarios with dedicated sections or steps.

Use information from an Architect flow

If you want to use information from a flow in the guide, create input variables for handling such information. For example, if a name is collected in the flow, you can start the guide with the following task:

- Say "Hello variable_contact_name"

If you want to use the information to determine the path the guide follows, you can use an If:Then:Else command. This command is important to ensure that the AI agent reliably knows what you want it to do with the information.

- If account == checking
- Then Go to Checking Account
- Else Go To Account Type

In this scenario, if a variable was not set in the flow, you must refer to it as None in the guide.

- If account_type == None
- Then Go to Get Account
- Else Go To Balance

Test and refine with preview

• Test your guide using the built-in preview widget.
• Test your guide by connecting it to an Architect bot flow and refine your instructions until you are satisfied with your AI agent’s performance. For more information, see Test and troubleshoot guides.