[dead]
My minute-by-minute response to the LiteLLM malware attack
161–170 of 194 posts
Re: My minute-by-minute response to the LiteLLM malware attack
#162If it weren't for the 11k process fork bomb, I wonder how much longer it would have taken for folks to notice and cut this off.
This generation's internet worm?
Re: My minute-by-minute response to the LiteLLM malware attack
#163Earlier quoted context omitted.
"got" can be read as "indirectly, via a blog post, which I think they reacted to"
I've updated the timeline to clarify I did in fact email them. I’m not yet at the point of having Claude write my emails for me, in fact it was my first one sent since joining the company 10 months ago!
Wait, what? You sent a single email being in a company for ten months?? Or was it the first external email?
Re: My minute-by-minute response to the LiteLLM malware attack
#164At this point I'd highly recommend everyone to think twice before introducing any dependencies especially from untrusted sources. If you have to interact with many APIs maybe use a proxy instead, or roll your own.
LiteLLM isn't a good choice for a proxy in any case. It introduces a lot of lag amd latency and the features are often half baked. To me, it looks like a vibecoded application without a product owner. And the code itself isn't very organized either. I evaluated it for a project a few months ago and will never use it for anything production.
Theres a few much better alternatives out there.
Re: My minute-by-minute response to the LiteLLM malware attack
#165I have a hard time believing that Claude instantly figured out this was malware... I've fed it obfuscated JavaScript before, and it couldn't figure it out... and then there was the time I tried to teach it nftables... whooo boy...
If you read the transcript it repeatedly made the incorrect assertion (hallucinated) that it’s totally normal for Claude Code to use Base64 armoring.
It’s not surprising it can “read” Base64 though; such was demonstrated back in GPT-3 days. Nontrivial obfuscation might not be one-shotted, but Claude has access to a code interpreter and can certainly extract and step through the decoder routine itself as a malware analyst would.
nftables is a different problem though. It’s apparent that if something isn’t well understood—i.e, there are tons of badly-formed examples on StackExchange—LLMs will fail to learn it too. I’ve seen this with things as “simple” as Bash string interpolation rules like ${var:+blah}. More often than not I’m humbled when I think I’ll learn it better and then find myself swearing at poorly-written documentation and patently false Q&A advice.
Re: My minute-by-minute response to the LiteLLM malware attack
#166What stands out to me the most:
> Blog post written, PR'd, and merged in under 3 minutes.
It's close to or even faster than the time it takes me to read it. I'm struggling to put into words how that makes me feel, but it's not a good feeling.
Re: My minute-by-minute response to the LiteLLM malware attack
#167If it weren't for the 11k process fork bomb, I wonder how much longer it would have taken for folks to notice and cut this off.
[dead]
Re: My minute-by-minute response to the LiteLLM malware attack
#168I have a hard time believing that Claude instantly figured out this was malware... I've fed it obfuscated JavaScript before, and it couldn't figure it out... and then there was the time I tried to teach it nftables... whooo boy...
> and then there was the time I tried to teach it nftables
I think this deserves a short story!
Re: My minute-by-minute response to the LiteLLM malware attack
#169I have a hard time believing that Claude instantly figured out this was malware... I've fed it obfuscated JavaScript before, and it couldn't figure it out... and then there was the time I tried to teach it nftables... whooo boy...
> and then there was the time I tried to teach it nftables I think this deserves a short story!
My home router is apparently special because I want a private dmz and inbound blocklists... firewalld couldn't do what I wanted, so I tried nftables, and that went down a rabbit hole that made me wish iptables and ipset weren't being replaced.
Re: My minute-by-minute response to the LiteLLM malware attack
#170Consider this your call to write native software. There is yet to be a supply chain attack on libc
Only because C code presents so many juicy security holes by default that it's completely unnecessary to subvert the projects to add them.