What would be the difficulty level for it to just read the machine code; are these models heavily relying on human language for clues?

▲

dnautics 4 minutes ago | parent | next [-]

I have had Claude read usbpcap to reverse engineer an industrial digital camera link. It was like pulling teeth but I got it done (I would not have been able to do it alone)

▲

wongarsu 4 hours ago | parent | prev | next [-]

Reasoning on pure machine code or disassembly is still hit and miss. For better results you can run the binary through a disassembler, then ask an llm to turn that into an equivalent c program, then ask it to work on that. But some of the subtleties might get lost in translation

▲

orwin 3 hours ago | parent [-]

If you put codex in Xhigh and allow it access to tools, it will take an hour but it will eventually give you back quality recompiled code, with the same issues the original had (here quality means readable)

▲

bryancoxwell 3 hours ago | parent [-]

I had a bit of a pain of a time trying to get Claude to work with ghidra. What you’re describing seems like a better alternative, would you agree?

	▲	skywal_l 2 hours ago \| parent [-]
		You can tweak the current Ghidra MCP to work in headless mode. It makes things much easier.

▲

lynx97 3 hours ago | parent | prev [-]

It will have to use a disassembler, or write one. I recently casually asked gpt-5.4 to translate the content of a MIDI file to a custom sound programming language. It just wrote a one-shot MIDI parser in Python, grabbed the data, and basically did a perfect translation at first try. Nice.

	▲	StilesCrisis 3 hours ago \| parent [-]
		I've seen Claude do similar things for image files. Don't have PNG parsing utilities installed? No worries, it'll just synthesize a Python script to decode the image directly.