Modification calls with Azure AI Agent Function is causing Security Alerts

Question

Trying this below POC with a function which can extract temperature data and print in Console: https://learn.microsoft.com/en-us/azure/ai-services/agents/how-to/tools/function-calling?tabs=csharp&pivots=code-example

This is very strange. I created multiple AI agent functions to retrieve data, and my code was working smoothly. However, when I created a function to submit data, I started receiving security alerts and malware protection warnings indicating that it was trying to access my project DLL. Can someone check and confirm the issue? I understand that running Visual Studio in admin mode might resolve this, but it shouldn't be necessary. Since I can run the function in normal mode, the function related to modification (which isn't actually modifying anything, as I've hardcoded all the values) should also be able to run.

Below is the screenshot regarding which no security alert occurred.

Below is the code which is throwing security alert. And yes if I keep the property Location, the alert is not coming. Apart from Location, it is throwing Malware alert.

Below is the security threat error message and build error.

Answer 1

Hi Deepankar,

Welcome to Microsoft Q&A and thank you for posting your question here.

Without proper code review this might be hard for a better solution.

Regarding the code sample, nothing wrong so far: https://learn.microsoft.com/en-us/azure/ai-services/agents/how-to/tools/function-calling?tabs=csharp&pivots=code-example

• Sometimes, functions that modify data or interact with certain resources require elevated permissions - https://learn.microsoft.com/en-us/azure/ai-services/agents/how-to/tools/azure-functions
• Ensure that your Visual Studio settings and project configurations are correctly set up to avoid unnecessary security scans - https://learn.microsoft.com/en-us/visualstudio/ide/user-permissions-and-visual-studio?view=vs-2022
• Also, consider adding exclusions for your project files and directories in your security software. This can prevent the security software from scanning these files and triggering false positives.
• You can check the detailed logs and alerts from your security software to provide more context on why the alerts are being triggered. Look for specific details about the DLL access that is causing the alert.

I hope this is helpful! Do not hesitate to let me know if you have any other questions or clarifications.

---

Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful.