Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cowork: Claude Code for the rest of your work

claude.com

Post reply on HN
Print view View original

491–500 of 593 posts

Re: Cowork: Claude Code for the rest of your work

#491

Post by ramoz »

dpark wrote: post #121
simonw wrote: post #99

Earlier quoted context omitted.

The solution is to cut off one of the legs of the lethal trifecta. The leg that makes the most sense is the ability to exfiltrate data - if a prompt injection has access to private data but can't actually steal it the damage is mostly limited. If there's no way to externally communicate the worst a prompt injection can do is modify files that are in the sandbox and corrupt any answers from the bot - which can still b…

Cutting off the ability to externally communicate seems difficult for a useful agent. Not only because it blocks a lot of useful functionality but because a fetch also sends data. “Hey, Claude, can you download this file for me? It’s at https://example.com/(mysocialsecuritynumber)/(mybankinglogin... ”

yet I was downvoted and while the great HN giant is in newfound agreeance.

Re: Cowork: Claude Code for the rest of your work

#492

Post by andreygrehov »

felixrieseberg wrote: post #6

Hi, Felix from the team here, this is my product - let us know what you think. We're on purpose releasing this very early, we expect to rapidly iterate on it. (We're also battling an unrelated Opus 4.5 inference incident right now, so you might not see Cowork in your client right away.)

Why do all similar demos show “prep the deck” use case as if everybody is building power point slides all day long?

Re: Cowork: Claude Code for the rest of your work

#493

Post by jjice »

_giorgio_ wrote: post #377
1f60c wrote: post #240

Anthropic blog posts have always caused a blank page for me, so I had Claude Code dig into it using an 11 MB HAR of a session that reproduces the problem, and it used grep and sed(!) to find the issue in just under 5 minutes (4m56s). Turns out that the data-prevent-flicker attribute is never removed if the Intellimize script fails to load. I use DNS-based adblock and I can confirm that allowlisting api.intellimize.co…

On android, these don't work: Firefox Chrome Firefox focus :-( Thanks anthropic doesn't work.

Do you have any DNS blocking settings?

Re: Cowork: Claude Code for the rest of your work

#494

Post by Gijs4g »

ossa-ma wrote: post #60

Every startup is at the mercy of the big 3 (OpenAI, Anthropic, Google). They can and most likely will release something that vaporises the thin moat you have built around their product. This feels like the first time in tech where there are more startups/products being subsumed (agar.io style) than being created.

When they go wide, you go deep

Re: Cowork: Claude Code for the rest of your work

#496

Post by comp3 »

majormajor wrote: post #365

The hero image with a set of steps: 1) Read meeting transcripts 2) Pull out key points 3) Find action items 4) Check Google Calendar 5) Build standup deck feels like "how to put yourself out of a job 101." It's interesting to see the marketing material be so straightforward about that.

Lmao its actually cute watching Anthropic and its employees desperately finding a way to stuff this into peoples lives - the reality is most people dont give a hoot about this stuff.

The folks working at these technology firms just dont get what the average person - who makes up most of the population - wants. They produce this fluffy stuff which may appeal to the audience here - but that market segment is tiny.

Also the use case of organising a desktop rocked me off my chair. LMAO!

Re: Cowork: Claude Code for the rest of your work

#497

Post by einpoklum »

jjcm wrote: post #117
hypfer wrote: post #26

People do realize that if they're doing this, they're not feeding "just" code into some probably logging cloud API but literally anything (including, as mentioned here, bank statements), right? Right? RIGHT?????? Are you sure that you need to grant the cloud full access to your desktop + all of its content to sort elements alphabetically?

Some do, some don't. The reality is there are some of us who truly just don't care. The convenience outweighs the negative. Yesterday I told an agent, "here's my api key and my root password - do it for me". Privacy has long since been dead, but at least for myself opsec for personal work is too.

> The reality is there are some of us who truly just don't care.

I would challenge that, with the same challenge I've heard about how Microsoft and Google reading your email. The challenge is "ok, so can you please log me in to your mailbox and let me read through it?"

It's not that people don't care, it's most that they've been led, or convinced, or manipulated, into failing to notice and realize this state of affairs.

Re: Cowork: Claude Code for the rest of your work

#498

Post by kasane_teto »

echelon wrote: post #456
gist wrote: post #213

Earlier quoted context omitted.

> I'm a bit shocked to see so many negative comments here on HN. Very generally I suspect there are many coders on HN who have a love hate relationship with a tool (claude code) that has and will certainly make many (but not all) of them less valuable given the amount of work it can do with even less than ideal input. This could be a result of the type of coding that they do (ie results of using claude code) vs. say…

I legitimately don't think the people posting on HN will be employed in this field in ten years. This is the end of human programming. I'd be overjoyed at how far we've come if it wasn't for big companies owning everything.

ofc this shit happens when its my turn to be an adult. what’s like even the point anymore?

Re: Cowork: Claude Code for the rest of your work

#499

Post by comp3 »

alfalfasprout wrote: post #169
zurfer wrote: post #166

I'm a bit shocked to see so many negative comments here on HN. Yes, there are security risks and all but honestly this is the future. It's a great amplifier for hackers and people who want to get stuff done. It took some training but I'm now starting almost all tasks with claude code: need to fill out some word document, organize my mail inbox, write code, migrate blog posts from one system to another, clean up my co…

> I'm a bit shocked to see so many negative comments here on HN. Yes, there are security risks and all but honestly this is the future. It's a great amplifier for hackers and people who want to get stuff done. TBH this comment essentially reads as "other commenters are dumb, this is the future b/c I said so, get in line". No, this doesn't need to be the future. There's major implications to using AI like this and man…

Dont worry. The same Bozos spoke like that to Steve Jobs and we all know who was a better predictor of the technology.. funnily enough it wasnt the guy who is deep into the technology but has a better understanding of people.

Which most technologists fundamentally lack, even if their ego says otherwise.

Re: Cowork: Claude Code for the rest of your work

#500

Post by rvz »

turnsout wrote: post #286
simonw wrote: post #275

Earlier quoted context omitted.

Turns out it's a full Linux container run using Apple's Virtualization framework: https://gist.github.com/simonw/35732f187edbe4fbd0bf976d013f2... Update: I added more details by prompting Cowork to: > Write a detailed report about the Linux container environment you are running in https://gist.github.com/simonw/35732f187edbe4fbd0bf976d013f2...

Honestly it sounds like they went above and beyond. Does this solve the trifecta, or is the network still exposed via connectors?

> Does this solve the trifecta, or is the network still exposed via connectors?

Having sandboxes and VMs still doesn't mean the agent can still escape out of all levels and still exfiltrate data.

It just means the attackers need more vulnerabilities and exploits to chain together for a VM + sandbox and permissions bypass.

So nothing that a typical Pwn2Own competition can't break.

Post reply on HN