I wanted to share a project I built in a strange but fun pair-programming trip with a large language model. The idea was to create my own automated "First Officer"—a tool that handles the tactical grunt work of finding low-hanging fruit vulnerabilities while I focus on the strategic, human-led parts of a security assessment.
The result is Nightcrawler, an open-source CLI proxy built on Python & mitmproxy.
As you browse a target, it passively finds insecure headers, outdated JS, and JWTs, while its active scanners autonomously test every discovered link and form for XSS, SQLi, Directory Traversal, and more.
The development was a real lesson in the "Picard/Riker" model of human-AI collaboration. I'd give the strategic orders ("We need to detect Stored XSS"), and the LLM would execute the tactical implementation (and occasionally "hallucinate," requiring my correction).
The tool is still in beta. I'd love to get your feedback, bug reports, or ideas on what to build next.
Thanks for checking it out!