Set rules for your AI agent

Rules are essential guidelines that help shape the behavior of your AI agent, ensuring it provides consistent, accurate, and appropriate responses during interactions. By establishing clear rules, you can control how your agent communicates, handles sensitive topics, and maintains professionalism.

What Are Rules?

Rules in MagicBlocks define specific behaviors or constraints that your AI agents must follow during conversations. These rules ensure that the agent adheres to your desired standards for communication, avoiding errors, inappropriate language, or misleading information. They are essentially a set of instructions that guide your AI agent on what to do, and what not to do, when engaging with users.

Why Are Rules Important?

Setting up rules for your AI agent is important for several reasons:

Consistency: Rules help maintain a consistent tone and approach throughout all interactions.

Professionalism: Ensures that the AI agent does not use inappropriate language, spread misinformation, or engage in offensive behavior.

Trustworthiness: Prevents the agent from providing inaccurate or misleading information, thereby building user trust.

Compliance: Ensures the AI adheres to legal, ethical, and company policies by avoiding sensitive topics or biased content.

User Satisfaction: Helps manage difficult situations, such as responding calmly to an angry user, thus improving the overall user experience.

How to Set Up Rules in the Studio

Follow these steps to set up rules for your AI agent:

Navigate to Agents: Start by accessing "Agents" section from the dashboard panel.

Select Your AI Agent: Choose the AI agent you want to configure and go into its settings.

Go to one of your Journey Blocks and click Guardians Section:

Select one of your Journey Blocks.

Click Guardians and head to Rules section.

Write your rules here (Play Now)

Write Your Rules:

Set any rules that the AI must follow in each interaction.

Examples of Rules

Here are some example of rules to ensure your AI agent behaves appropriately and meets your standards. Keep it simple!

• Never make anything up
• Don't use derogatory language and don't offend users
• Promote inclusivity and avoid bias
• Stay neutral on religious or political or controversial views
• Don't be misleading - avoid spreading false or misleading information, and always fact-check responses
• Always be polite, regardless of the user's tone
• Avoid arguments and calmly respond if someone is angry
• Provide clear responses, don't use confusing language or jargon
• Keep conversations as concise as possible, while still making sense
• Promote accessibility by using simple and clear language
• Always acknowledge the user's responses, even if you don't have an answer
• Always be transparent about being an AI - don't pretend to be a real human
• Avoid sensitive topics unless specifically instructed to handle them
• Follow legal guidelines for {{user.country}}
• Take the time to customize rules that suit your needs, and remember to review them periodically to adapt to any new requirements or changing user expectations.