An AI agent in the fuzzing loop: re-finding CVE-2017-11729 in libming
Re-finding CVE-2017-11729, a heap over-read in libming's SWF decompiler, with an AI coding agent running the fuzzing. Here's where it tried to lie to me.
#llm
4 posts
Check reachability before you fuzz for a CVE
When I pick a known CVE to re-find by fuzzing, the question that matters isn't how bad the bug is or who found it. It's how many conditions a fuzzer has…
herdr + omp: my terminal setup for agent-driven security work
Two tools do most of my agent work now. omp (oh my pi, a fork of Mario Zechner's Pi) is the coding agent. herdr is the terminal it lives in. They solve…
JS file enumeration, four years later: recon with LLMs
Follow-up to my 2022 post: how to collect an SPA's JavaScript, cut it down, and let an LLM read it without burning tokens or trusting a word it says.