large-language-models

Here is a walkthrough of our solutions to two C/C++ security challenges from the new C/C++ chapter in the Testing Handbook: a Linux ping program with an inet_ntoa global buffer gotcha and command injection, and a Windows driver with registry type confusion bugs that can escalate from local DoS to a kernel write primitive.

We released a Claude plugin that uses LLMs to annotate code with dimensional types and mechanically detect mismatches, achieving 93% recall versus 50% for baseline prompts when tested against real audit findings.