LLM : Your Security Auditor
As I was researching about the code vulnerabilities in my 42 Wolfsburg/Berlin Webserver project, I found Large Language Model can be a good security auditor.
When tested out leveraging Cohere Command R+ and Google Gemini, I dove into my code, hoping to unearth hidden threats.
The Results? Eye-Opening 😱
The LLMs identified several potential security issues in my code, some of which aligned perfectly with the OWASP OWTF security risks!
How Generative AI can offer developers immense productivity boost in identifying vulnerabilities within their code. 🤓
Let’s look into the below result :
COHERE :
GEMINI :
From the above output one can see how Cohere and Gemini can realize the code vulnerabilities. One can leverage the prompt
“Help me find security vulnerability in the given code “
If you’re working in a company or somewhere where your code cannot be pushed directly in a large language model, one can use it to create a different pattern of code keeping the similar principle and find the security check in there.
If the above writing make you look forward to follow secure coding practices feel free to connect in linkedin / message me in slack. Also do share your findings below.
#42 #born2code #secureCoding #back2Reality
