There are a set of ongoing uses cases and best practices for Generative AI for infosec use cases.
Currently a script which takes a list of Languages e.g (Python, Java, JS, C# etc..), and for each language describe the following list of CWEs:
- Describe the vulnerability type
- Show an example of vulnerable code in that language
- Show an example of remediation
- Create a Semgrep rule to detect the type of vulnerability
- Create a CodeQL rule to detect the type of vulnerability
The goal is to show how you can write the structure of a long set of technical best practices, requirments and OpenAI can write the initial DRAFT of those best practices and techincal information quickly for people to then review and edit.
This is a weba application trained on the OWASP OpenCRE Infosec References and returns responces with the correct citations