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

Claude Code 2.0

npmjs.com

Post reply on HN
Print view View original

281–290 of 431 posts

Re: Claude Code 2.0

#281

Post by tuyiown »

athrowaway3z wrote: post #62
simonw wrote: post #49

Something I realized about this category of tool (I call them "terminal agents" but that already doesn't work now there's an official VS Code extension for this - maybe just "coding agents" instead) is that they're actually an interesting form of general agent . Claude Code, Codex CLI etc can effectively do anything that a human could do by typing commands into a computer. They're incredibly dangerous to use if you d…

They're only as dangerous as the capabilities you give them. I just created a `codex` and `claude` user on my Linux box and practically always run in yolo mode. I've not had a problem so far. Also, I think shellagent sounds cooler.

> They're only as dangerous as the capabilities you give them.

As long as the supply chain is safe and the data it accesses does not generate some kind of jail break.

It does read instructions from files on the file system, I pretty sure it's not complex to have it poison its prompt and make it suggest to build a program infected with malicious intent. It's just one copy pasta away from a prompt suggestion found on the internet.

Re: Claude Code 2.0

#282

Post by philipp-gayret »

polyrand wrote: post #260
simonw wrote: post #49

Something I realized about this category of tool (I call them "terminal agents" but that already doesn't work now there's an official VS Code extension for this - maybe just "coding agents" instead) is that they're actually an interesting form of general agent . Claude Code, Codex CLI etc can effectively do anything that a human could do by typing commands into a computer. They're incredibly dangerous to use if you d…

Instead of containers, which may not always be available, I'm experimenting with having control over the shell to whitelist the commands that the LLM can run [0]. Similar to an allow list, but configured outside the terminal agent. Also trying to make it easy to use the same technique in macOS and Linux [0]: https://ricardoanderegg.com/posts/control-shell-permissions-...

I really like this and we're doing a similar approach but instead using Claude Code hooks. What's really nice about this style of whitelisting is that you can provide context on what to do instead; Let's say if `terraform apply` is banned, you can tell it why and instruct it to only do `terraform plan`. Has been working amazing for me.

Re: Claude Code 2.0

#283

Post by user3939382 »

polyrand wrote: post #260
simonw wrote: post #49

Something I realized about this category of tool (I call them "terminal agents" but that already doesn't work now there's an official VS Code extension for this - maybe just "coding agents" instead) is that they're actually an interesting form of general agent . Claude Code, Codex CLI etc can effectively do anything that a human could do by typing commands into a computer. They're incredibly dangerous to use if you d…

Instead of containers, which may not always be available, I'm experimenting with having control over the shell to whitelist the commands that the LLM can run [0]. Similar to an allow list, but configured outside the terminal agent. Also trying to make it easy to use the same technique in macOS and Linux [0]: https://ricardoanderegg.com/posts/control-shell-permissions-...

You have to put them in the same ACL, chroot, whatever permission context for authorization you’d apply to any other user human or otherwise. For some resources it’s cumbersome to setup but anything else is a hope and a prayer.

Re: Claude Code 2.0

#284

Post by numpad0 »

fyi: for chatboxes that may take CJK inputs, you MUST use "shift+enter to send" pattern. There is a reason why most multinational chat/LLM app providers always do that instead of simple enter to send even for single-line chatboxes; because plain enter to send breaks input for CJK users.

Specifically, Input Method Editors needed for CJK inputs(esp. for C and J), to convert ambiguous semi-readable forms into proper readable text, use enter to finalize after candidates were iterated with spacebar. While IME engines don't interchange between different languages, I believe basically all of them roughly follow this pattern.

Unless you specifically wants to exclude CJK users, you have to either detect presence of IME and work with it so that enter do nothing to the app unless conditions are met. Switching to shift+enter works too.

1: https://github.com/anthropics/claude-code/issues/8405

2: https://www.youtube.com/watch?v=mY6cg7w2eQU

3: https://youtu.be/sYAnawy_VoA?feature=shared&t=282

4: https://www.youtube.com/watch?v=VmoeZ_W3WXo

Re: Claude Code 2.0

#286

Post by adastra22 »

SparkyMcUnicorn wrote: post #254
adastra22 wrote: post #215

Earlier quoted context omitted.

I have this same suspicion. Worse, there’s no way to opt out of giving a response.

If you turn off "Help improve Claude" you will never get this prompt (I never do). https://claude.ai/settings/data-privacy-controls

That should be how this works, but unfortunately not. I have that toggle switched off, but I still regularly get this prompt.

Re: Claude Code 2.0

#287

Post by monkeydust »

simonw wrote: post #49

Something I realized about this category of tool (I call them "terminal agents" but that already doesn't work now there's an official VS Code extension for this - maybe just "coding agents" instead) is that they're actually an interesting form of general agent . Claude Code, Codex CLI etc can effectively do anything that a human could do by typing commands into a computer. They're incredibly dangerous to use if you d…

Been starting to wonder if this marks a step change in UX - moving away from pretty well designed screens where designers labor over positioning of artifacts like buttons, user input dialogs and color palettes to a CLI! I cant imagine CLI will work for everything but for a lot of things, when powered by LLM they are incredible and yea equally dangerous at the same time for many reasons.

Re: Claude Code 2.0

#288

Post by stefan_ »

kelnos wrote: post #123
marckrn wrote: post #12

You can find the revamped prompt on github[1], or on twitter summarized by my bot[2]. [1] https://github.com/marckrenn/cc-mvp-prompts/compare/v1.0.128... [2] https://x.com/CCpromptChanges/status/1972709093874757976

> IMPORTANT: DO NOT ADD * ANY** COMMENTS unless asked* Interesting. This was in the old 1.x prompt, removed for 2.0. But CC would pretty much always add comments in 1.x, something I would never request, and would often have to tell it to stop doing (and it would still do it sometimes even after being told to stop).

Meanwhile they deleted the "do not add emojis" part. Look forward to all sorts of logging messages with emojis in them.

Re: Claude Code 2.0

#289

Post by budududuroiu »

simonw wrote: post #49

Something I realized about this category of tool (I call them "terminal agents" but that already doesn't work now there's an official VS Code extension for this - maybe just "coding agents" instead) is that they're actually an interesting form of general agent . Claude Code, Codex CLI etc can effectively do anything that a human could do by typing commands into a computer. They're incredibly dangerous to use if you d…

I’m experimenting with Nix shells for this tool isolation and whitelisting

Re: Claude Code 2.0

#290

Post by athrowaway3z »

globular-toast wrote: post #257
athrowaway3z wrote: post #62

Earlier quoted context omitted.

They're only as dangerous as the capabilities you give them. I just created a `codex` and `claude` user on my Linux box and practically always run in yolo mode. I've not had a problem so far. Also, I think shellagent sounds cooler.

I tried this but it's incredibly annoying as you'll get a mixture of file ownerships and permissions. Instead I run it in bubblewrap sandbox: https://blog.gpkb.org/posts/ai-agent-sandbox/

Set umask from 022 to 002 to give group members the same permissions as a user.
Post reply on HN