Please give a quick overview of your product feature request or feedback and note who in your org is affected by this issue.
It would be helpful if AI Agent (Advanced) Instructions could be targeted based on user attributes such as tags, user fields, or organization fields. This would allow the AI agent to provide different responses depending on customer context or entitlements. This primarily affects administrators configuring AI agents and customers interacting with them.
What problem do you see this solving?
Currently, instructions appear to apply globally and cannot easily be adjusted based on customer attributes. Allowing conditional targeting would enable more personalized and accurate AI responses.
When was the last time you were affected by this lack of functionality, or specific tool? What happened? How often does this problem occur and how does this impact your business?
This limitation appears when configuring AI agents to handle different types of customers with varying support entitlements. Without conditional instructions, it becomes difficult to ensure the AI agent provides the correct guidance depending on the customer context. This can lead to inconsistent experiences where customers may receive instructions that do not match their available support options. The issue occurs whenever organizations need to differentiate responses for different customer groups.
Are you currently using a workaround to solve this problem?
Workarounds typically involve using segments, search rules, or custom logic outside of the Instructions feature. However, these approaches do not provide the same level of direct control within the AI instruction configuration.
What would be your ideal solution to this problem? How would it work or function?
The ideal solution would allow conditional logic within AI Agent Instructions, similar to how segments work. Instructions could be applied only when certain conditions are met, such as specific user tags, user fields, or organization attributes.
Thank you for taking the time to provide us with your feedback. This has been logged for our PM team to review. For others who may be interested in this feature request, please add your support by upvoting this post and/or adding your use case to the comments below. Thank you again!