shitter

.onion

? Doesn’t the code exec happen upon merely launching Cursor against a repository, without giving an agent any tasks? That’s clearly an issue at the Cursor application level, not some inevitable risk caused by the non-deterministic nature of LLMs. You can’t use the latter fact to excuse the former mistake

Not to mention that Cursor has an agent permissions model that this presumably sidesteps!

I would file a CVE for any program that places untrusted content into PATH and invokes non-fully qualified executable names - not for the shell.

Yeah, but you can easily mitigate it by searching for the real git in known system locations and using whatever you find there (or allowing the user to configure the path). I believe that's how VSCode does it

Since there are no approval dialogs, it sounds like that doesn't even come into play here. That is the "gate" (to use the AI parlance) that Microsoft places on code execution in workspaces, though, and I would expect Cursor to at minimum fix this to only execute git.exe in trusted workspaces.

The user's code folder? You mean the code I frequently pull from untrusted sources, unlike my .bashrc? Opening a GitHub project for review should not mean arbitrary code execution.

Of course, that ship has long sailed, for all major IDEs. Heck, VSCode SSH and devcontainer remotes allow RCE by design.

And what'll the prompt say? "Do you want to run git.exe?"? I'll probably assume Cursor needs to run git but permissions got messed up somewhere and click right through that.

I haven't used Windows in a while so pardon if I'm missing something.

Temporary in the sense that it's a stop or two away from getting laid off.

This comment betrays a misunderstanding of history and of what fascism is and doesn't even engage with the article's points. A totalitarian fascist state does not come into being all at once; it emerges in steps that tend towards that outcome, steps which the article discusses in detail. Furthermore, that nascent fascism can be defeated through electoral means does not preclude it from being fascism.

An I.C.U. nurse shot by federal agents was an American citizen with no criminal record, the city police chief said. A New York Times video analysis shows he was holding a phone, not a gun.

https://www.nytimes.com/live/2026/us/minneapolis-shooting-ic...

Call me a relic of a bygone era of seriousness and decency, but I don't think the White House should be in the business of mocking individuals online, either. But I know we're well past that point.

Why not? If it's high-performance, it's fine.

It shows no such thing. The officer got out of the way and was fine afterwards in both videos. This is a man who was angry and possibly traumatized by a prior incident involving a car. He should not have been allowed to carry a weapon

JD Vance, man. What a mendacious piece of shit.

Actually, the case law seems pretty clear in the other direction. Deadly force is not justified once an officer has evaded a vehicle that might hit them.

"Cases where police moved out of the way of moving vehicle and exercised deadly force": https://old.reddit.com/r/law/comments/1q6o4d0/comment/ny95v2...

I share the author's perspective that LLMs are not fun for programming. I don't use them to generate code, save for small snippets to demonstrate some concept or do something rote that I wouldn't enjoy writing myself.

However - and maybe I'm just an easily entertained simpleton - I find them really fun for exploring those random, not trivially Google-able questions that pop into my head on a daily basis, technical and otherwise. Most of my chats with ChatGPT begin with questions of this form. I keep my critical thinking cap on during the dialogue and always verify the output if it's to be used for anything serious, but I'd be lying if I said I didn't enjoy the process.

I'm a naturally socially awkward person, due partly to personality and partly to social anxiety disorder. I don't think I have the social intelligence/agility to pull off half of these maneuvers. Just reading about the author playing different characters with tables and imagining myself in that position made me want to throw up, I'd fuck it up so badly.

You sound a little like me. I wasn't aware people thought like that until my partner told me her thoughts took the form of a constant stream of well-formed English. My default mode of thinking isn't natural language (though I can force myself to think this way, it's laborious, as the article mentions), nor images (I struggle with visualization), but more like abstract sequences of both logical connections and intuitive feelings.

This is part of it, but everyone to some degree has discomfort with being disliked and will do things to avoid it. At least in my experience, social anxiety is much more about the cognitive distortions that convince you others dislike you, when they may in fact be neutral or even have a positive view of you.

Just as one example, when I'm interacting with someone who I haven't reached a certain level of comfortability with, I'm highly aware of and sensitive to their reactions to me in terms of what they're saying, their tone, their micro facial expressions, etc., and I perceive any small negative reaction as a sign that they don't like me. This usually isn't true! But it ironically has the effect of inducing self-sabotaging avoidant behaviors in me, such as over-censoring of what I say and just general awkwardness around them, which makes it much more likely they will end up disliking me.

https://archive.is/CecEx

https://www.washingtonpost.com/immigration/2025/04/26/us-cit...

Only took 9 days for you to lose this bet.

You probably shouldn't accuse others of being ungrounded in reality if you don't keep up with reality.

Regardless of what it is called, it is unconstitutional and illegal; however, that does not seem to matter to this administration. Trump is currently defying a 9-0 SCOTUS ruling that he facilitate the return of Abrego Garcia to the US. His administration has even admitted Garcia was deported in error but still isn't doing anything about it. Now we have ICE holding provable US citizens. Maybe with that context you can understand why some might be concerned that citizens will be "accidentally" disappeared.

I imagine the conversation between the CEO and his reports included something about "it's no biggie, the passwords were hashed using bcrypt, that's like irreversible encryption" without contextualizing that and mentioning that plaintext auth tokens were also exposed.

https://www.mysql.com/industry/government/

It's not a lie. "I don’t think stupid social media activity should ruin a kid’s life" does two things: characterizes endorsements of racism and eugenics policies as "stupid social media activity", and asserts that exposure of such activity should not lead to someone losing their job. The part about not rewarding journalists who "destroy" people does not exist in a vacuum away from his downplaying of Elez's abhorrent racial views. If journalists had revealed Elez, for example, was a secret left-wing antifa supporter on Bluesky, I think we can reasonably doubt Vance's reaction would have been the same.

Edit: Now that you've brought Vance's tweet into focus, it's also interesting that he does not think Elez is currently a "bad dude" when he's expressly stated a desire to normalize hatred against an entire ethnicity

For reference: JD Vance said that another DOGE staffer, who resigned after it was reported that he'd publicly come out in favor of racism, eugenics, and normalizing hate against Indian people on Twitter, should be rehired because it was just "stupid social media activity".

However any agent based system exposing an arbitrary API is suddenly a much bigger attack and risk surface area than the well trod (and still fraught) path of emulating a terminal over ssh.

I can see how this increases local (to the remote system) attack surface, but as long as the agent has the same OS privileges as the user logged in over SSH, what extra remote risk does this introduce?

It works on my Mac and Linux box!

  echo "Phyllis Fong, who was investigating Elon Musk's brain implant startup Neuralink, "forcefully removed from office" after refusing termination order" | sed -E "s/Phyllis Fong, who was investigating (.*) startup Neuralink,(.*) from office(.*) refusing/\1\2\3/"

This is mitigated if your initramfs + kernel are measured into the TPM, right?

Edit: never mind, I think it's still vulnerable.