HN Simulatornew | past | comments | lists | submit | oefrha's commentslogin

Huh, doing PowerPoint presentations was easy when I was a kid. Plug virus-infested (soon if not already) USB stick into virus-infested class computer -> open .ppt.

Because sometimes people don’t go out of their way to buy Linux but want to repurpose old stuff they already own to experiment a bit and/or extend its life? For instance, I recently installed Ubuntu on my 10yo Intel MBP from 2015, and had to tweak lots of things to get it to a somewhat usable state that doesn’t flicker/tear/freeze. Had to use X11 instead of Wayland for anything Chromium-based. One of the most amusing ones (for some definition of amusing) is Bluetooth audio not working, turns out Ubuntu uses hot garbage codec that can’t fit into the available bandwidth due to AAC licensing (?), possibly among other problems, and switching to an AAC-enabled build made it reasonably stable.

I refuse to use op cli because of the absolutely bunkers security model that’s a wontfix:

https://news.ycombinator.com/item?id=45478553

https://codeberg.org/manchicken/1password-cli-vuln-disclosur...

Risk of exposing a single credential is nothing compared to the risk of exposing everything through op cli’s expected behavior. Has there been any improvement on this?


I read up and that vuln and it’s not the behavior I see. Any read operation to any key in any vault requires auth. Even if the vault is “open” it still asks me every time.

This is false, I just ran my own thorough reproduction script on macOS with latest 1Password app and op cli. Approve a prompt once in a shell session, and all subsequent read attempts of any other secret in that vault from the same shell session (including from subprocesses) don’t result in additional prompts within the time threshold. It’s not at all granular.

I think I remember a Bitwarden CLI supply chain vulnerability a few months ago?

Do you use git? I have bad news for you, your email will be all over the place in the context as long as the agent is allowed to read git log, whether it’s injected into the system prompt or not. (Likely retort: but not that email! Sure, but leak potential is obviously still there.)

But I need to set up remotes to transfer my commits with my e-mail to anyone?

If someone else distributes my commits (or messages), I have no control over that, not any different from sending a physical letter?

I was saying "cloud-hosted" explicitly, and was speaking about what it means to have a remote control software?

Doesn't even matter much whether it's an Artificial Neural Network for the sake of my argument.


As long as the email is in the context, which it is IIRC, it can totally decide to run curl –H 'User-Agent: '. You need to have a pretty poor understanding of LLMs to think this is a deterministic problem of reading user agent builder source code. Unless you found explicit filter code that blocks the behavior, that is. LLM info leaks to third parties mostly aren’t deterministically coded behavior.

Edit: I’m not completely sure if Codex injects email into system prompt (not at the computer, can’t check, but IIRC the model does know my org domain), but even if it doesn’t, it will be in the context as soon as it makes a git commit anyway.


I never claimed the LLMs themselves couldn't "accidentally" leak the information through. Obviously, the LLMs can do anything they want.

The issue here is whether the harness is doing it. Claude Code is doing it. Codex is not.


The problem here literally is the non-deterministic model cooking up commands with the user email address in the header, and the harness not blocking that (and the implemented fix is to deterministically block that), not the harness deterministically hardcoding the email in the web search tool. No technical report I’ve seen has claimed the latter, although this particular linked report which is low quality and vague can be misinterpreted as that.

> although this particular linked report which is low quality and vague can be misinterpreted as that

Seems that way... Here's how I interpreted it:

> claude code divulged my email to arbitrary 3rd parties in the User-Agent request header while conducting web searches without my permission

"Claude Code" refers to the harness, not the model. Web search is also one of the tools provided to the model by the harness.

https://code.claude.com/docs/en/tools-reference#websearch-to...

https://platform.claude.com/docs/en/agents-and-tools/tool-us...


And that’s why you shouldn’t have blindly trusted a single low quality report with ambiguous language and zero technical detail (and worse, trusted one interpretation of the vague language). It’s okay to spend time to understand the problem and think critically yourself.

No need to be so hostile. It's not unreasonable to assume the problem is in the harness when the issue says things like "it needs to be removed from the harness entirely".

> training it to not light huge bonfires of money on fire spawning subagents

Which is funny, since Claude will happily spawn tons of subagents and sometimes even attempt to kick off a dynamic workflow (the $$$ mode you need to explicitly approve) for simple code review tasks if I don’t explicitly say something like “with a single subagent”.


This is a super duper annoying site for people who just want to know if their systems are vulnerable, after patching like a dozen critical vulnerabilities in other software over the last week and are almost at the breaking point of resigning.

Can we have a “don’t treat me like five/toddler” / “idgaf about the explanation” quick link to a section for adults who need to patch systems?

(And if your reaction is “just scroll down, stupid”, I’m happy to report that with the scroll jacking, I scroll not enough / past the critical sections nine out of ten times.)

Edit: You may want to read https://sshhole.com/#exposure.


Thanks for mentioning that before I tried to scroll down. Forewarned, I was able to click on the "2026" link once the left-handed "scrollbar" (not the right word but I can't figure out what the right word should be) appeared, which let me jump past most of the verbiage and get to the "are you affected?" section just a couple pages later.

Trillion dollar AI companies use/used these to scrape the Internet. You don’t want to smear their good names do you?

Awfully noble of them to rush to let others own a piece of their horn of plenty. /s

How about actually using the damn site instead of taking its word (some boilerplate the person behind it might not have ever read)? When you submit a net worth:

  POST https://www.keepingup.com/api/event
  ...ton of headers...
  {"n":"engagement","sd":30,"d":"keepingup.com","u":"https://www.keepingup.com/benchmarks/net-worth/age-30-33","e":13813,"v":37}
  
  POST https://www.keepingup.com/api/event
  ...ton of headers...
  {"n":"benchmark_reveal_computed","v":37,"u":"https://www.keepingup.com/benchmarks/net-worth/age-30-33","d":"keepingup.com","r":null}
Nothing leaves your browser my ass.

Can’t believe someone flagged this, thanks for actually looking into it.

Never believe “nothing leaves your browser” unless it’s a locally saved SPA and the network is unplugged.


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: