Remediations play a pivotal role in effective client management, allowing organizations to proactively identify and resolve end-user issues. Additionally, they serve as a valuable tool for enforcing specific settings or configurations that may not be natively supported in Microsoft Intune. However, the process of crafting these scripts can often be intricate and time-consuming.

Imagine a solution where you can simply describe your desired configurations, and a tool generates the necessary scripts for you. If you find this idea appealing and are keen to explore such a solution, this blog is tailored to meet your exact needs.

Content

1. Content
2. Get the Tool
3. What is the GPT Remediation creator?
4. What are the prerequisites
5. How does the tool work
6. How to get access to Open AI Service
7. How it the data privacy when I use the Azure Open AI service
8. How to start the GPT Intune Device Troubleshooter
9. How you can contribute?

Get the Tool

https://github.com/JayRHa/Remediation-Creator

What is the GPT Remediation creator?

In the end, the process is remarkably straightforward. The tool generates a detection script and, if desired, a remediation script based on your provided description. It then seamlessly uploads these scripts to your tenant, streamlining the entire process.

What are the prerequisites

The prerequisites for getting started are minimal. All you need is a GPT-enabled subscription. Once you have this, you’ll need to create an Azure OpenAI Service deployment. After completing this step, you can easily add the relevant information to your ‘secrets.toml’ file. To do this, simply rename ‘secrets.toml.example‘ to ‘secrets.toml‘.

Another prerequisite is that python is installed on the system.

How does the tool work

1. User authentication: The process starts with user authentication, ensuring that only authorized users can upload scripts using the user token
2. User describes the purpose of the script: After authentication, the user provides a description of what the script is intended to do.
3. Building a prompt with instructions and the description to generate a detection script: This description, along with instructions, is used to build a prompt for GPT-3.5-turbo, which will then generate a detection script based on the input.
4. (Optional) Building an additional prompt with instructions, the detection script, and the description to generate a remediation script: Optionally, a further step involves building another prompt that includes the instructions, the detection script that was generated, and the description. This new prompt is for GPT-3.5-turbo to create a remediation script, presumably to address the issues the detection script identified.
5. Display at web page: The scripts or the output from GPT-3.5-turbo are then displayed on a web page.
6. User presses upload and script will add to tenant: Finally, the user can press an upload button, resulting in the script being added to the tenant.

How to get access to Open AI Service

Fill out the following request form to enable you subscription to deploy the OpenAi services.

How it the data privacy when I use the Azure Open AI service

For detailed information on privacy, please refer to this Microsoft documentation. In this scenario, no only personal information is transferred to the service. The only informaiton which is transfered is the script description provided by you.

How to start the GPT Intune Device Troubleshooter

The only thing what you have to do is to run the run.ps1 on Windows or the run.sh on Mac.

Alternativly you can run python -m streamlit run app.py in the terminal. Make sure that the packages in the requirement.txt are installed (streamlit, azure.identity)

How you can contribute?

If you have a ideas for improvements or for missing features as well as bugs, contact me via my blog, social media or open an issue on the repository with an description of your idea. You can also create merge request. I am happy for every contribution and feedback.