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

Claude Code can debug low-level cryptography

words.filippo.io

Post reply on HN
Print view View original

101–110 of 217 posts

Re: Claude Code can debug low-level cryptography

#101

Post by ewoodrich »

chatmasta wrote: post #81
danielbln wrote: post #78

Earlier quoted context omitted.

Claude Code is not a VSCode fork, it's a terminal CLI. It's a rather different interaction paradigm compared to your classical IDE (that said, you can absolutely run Claude Code inside a terminal inside VSCode).

Ah, I think I’m getting it confused with Cursor. So Claude Code is a terminal CLI for orchestrating a coding agent via prompts? That’s different than the initial releases of VSC copilot, but now VSC has “agent” mode that sounds a lot like this. It basically reduces the IDE to a diff viewer.

There is now also copilot-cli that's a clone of Claude Code and by default runs with Sonnet 4.5. I haven't spent much time with it on really complex stuff yet but it's a nice option to have available if you have a Copilot Business/Enterprise plan at work.

https://github.com/github/copilot-cli

Re: Claude Code can debug low-level cryptography

#102

Post by medvezhenok »

manquer wrote: post #93
mschulkind wrote: post #31

Earlier quoted context omitted.

One of my favorite ways to use LLM agents for coding is to have them write extensive documentation on whatever I'm about to dig in coding on. Pretty low stakes if the LLM makes a few mistakes. It's perhaps even a better place to start for skeptics.

I am not so sure. Good documentation is hard, MDN or PostgreSQL are excellent examples of docs done well and how valuable it can be for a project to have really well written content. LLMs can generate content but not really write , out of the box they tend to be quote verbose and generate a lot of proforma content. Perhaps with the right kind of prompts, a lot of editing and reviews, you can get them to good, but at…

Sure, but LLMs tend to be better at navigating around documentation (or source code when no documentation exists). In agentic mode, they can get me to the right part of the documentation (or the right of the source code, especially in unfamiliar codebases) much quicker than I could do it myself without help.

And I find that even the auto-generated stuff tends to go up at least a bit in terms of level of abstraction than staring at the code itself, and helps you more like a "sparknotes" version of the code, so that when you dig in yourself you have an outline/roadmap.

Re: Claude Code can debug low-level cryptography

#103

Post by faxmeyourcode »

chatmasta wrote: post #72
simonw wrote: post #41

Earlier quoted context omitted.

I think it might be that they've hit product-market fit. Developers find Claude Code extremely useful (once they figure out how to use it). Many developers subscribe to their $200/month plan. Assuming that's profitable (and I expect it is, since even for that much money it cuts off at a certain point to avoid over-use) Anthropic would be wise to spend a lot of money on marketing to try and grow their paying subscribe…

What makes it better than VSCode Co-pilot with Claude 4.5? I barely program these days since I switched to PM but I recently started using that and it seems pretty effective… why should I use a fork instead?

Copilot as a harness for the model is generic enough to work with every model. Claude sonnet, haiku, and opus are trained with and specifically for Claude code in mind.

Also, as a heavy user of both, there are small paper cuts that seriously add up with copilot. Things that are missing like sub agents. The options and requests for feedback that cc can give (interactive picker style instead of prompt based). Most annoyingly commands running in a new integrated vscode terminal instance and immediately mistakenly "finishing" even though execution has just begun.

It's just a better harness than copilot. You should give it a shot for a while and see how you like it! I'm not saying its the best for everybody. At the end of the day these issues turn into something like the old vi/emacs wars.

Not sponsored, just a heavy user of both. Claude code is not allowed at work, so we use copilot. I purchased cc for my side projects and pay for the $125/m plan for now.

Re: Claude Code can debug low-level cryptography

#104

Post by joshvm »

rtpg wrote: post #61
simonw wrote: post #11

Using coding agents to track down the root cause of bugs like this works really well: > Three out of three one-shot debugging hits with no help is extremely impressive. Importantly, there is no need to trust the LLM or review its output when its job is just saving me an hour or two by telling me where the bug is, for me to reason about it and fix it. The approach described here could also be a good way for LLM-skepti…

I understand the pitch here ("it finds bugs! it's basically all upside because worst case there's no output anyways"), but I'm finding some of these agents to be ... uhhh... kind of agressive at trying to find the solution and end up missing the forest for the trees. And there's some "oh you should fix this" stuff which, while sometimes isn't _wrong_, is completely besides the point. The end result being these robots…

When models start to forage around in the weeds, it's a good idea to restart the session and add more information to the prompt for what it should ignore or assume. For example in ML projects, Claude gets very worried that datasets aren't available or are perhaps responsible. Usually if you tell it outright where you suspect the bug to be (or straight up tell it, even if you're unsure) it will focus on that. Or, make it give you a list of concerns and ask you which are valid.

I've found that having local clones of large library repos (or telling it to look in the environment for packages) is far more effective than relying on built-in knowledge or lousy web search. It can also use ast-grep on those. For some reason the agent frameworks are still terrible about looking up references in a sane way (where in an IDE you would simply go to declaration).

Re: Claude Code can debug low-level cryptography

#105

Post by ewoodrich »

cmrdporcupine wrote: post #8
Frannky wrote: post #6

Earlier quoted context omitted.

I started with Claude Code, realized it was too much money for every message, then switched to Gemini CLI, then Qwen. Probably Claude Code is better, but I don't need it since I can solve my problems without it.

Try what I've done: use the Claude Code tool but point your ANTHROPIC_URL at a DeepSeek API membership. It's like 1/10th the cost, and about 2/3rds the intelligence. Sometimes I can't really tell.

Also there is a fixed price Claude Code plan for GLM 4.6 from z.ai, I pay for the cheapest ($6/mo) as an alternate/fallback for Claude Code and Codex. I've been surprised by how similar in capabilities all three of them are, definitely not seeing a big CLI agent moat...

GLM is maybe slightly weaker on average but on the other hand it's also solved problems where both CC and Codex got stuck in endless failure loops so for the price it's nice to have in my back pocket. I also see some tool use failures sometimes that it always works around which I'm guessing are due to slight differences with Claude.

https://z.ai/subscribe

Re: Claude Code can debug low-level cryptography

#106

Post by Terr_ »

theptip wrote: post #44
jack_tripper wrote: post #22

Earlier quoted context omitted.

>Have the coding agents do the work of digging around hunting down those frustratingly difficult bugs - don't have it write code on your behalf. Why? Bug hunting is more challenging and cognitive intensive than writing code.

Bug hunting tends to be interpolation, which LLMs are really good at. Writing code is often some extrapolation (or interpolating at a much more abstract level).

Reversed version: Prompting-up fresh code tends to be translation, which LLMs are really good at. Bug hunting is often some logical reasoning (or translating business-needs at a much more abstract level.)

Re: Claude Code can debug low-level cryptography

#107

Post by Wowfunhappy »

ori_b wrote: post #83
Wowfunhappy wrote: post #80

Earlier quoted context omitted.

Sometimes you hit a wall where something is simply outside of the LLM's ability to handle, and it's best to give up and do it yourself. Knowing when to give up may be the hardest part of coding with LLMs. Notably, these walls are never where I expect them to be—despite my best efforts, I can't find any sort of pattern. LLMs can find really tricky bugs and get completely stuck on relatively simple ones.

Doing it yourself is how you build and maintain the muscles to do it yourself. If you only do it yourself when the LLM fails, how will you maintain those muscles?

I agree, and I can actively feel myself slipping (and perhaps more critically, not learning new skills I would otherwise have been forced to learn). It's a big problem, but somewhat orthogonal to "what is the quickest way to solve the task currently in front of me."

Re: Claude Code can debug low-level cryptography

#108

Post by chatmasta »

faxmeyourcode wrote: post #103
chatmasta wrote: post #72

Earlier quoted context omitted.

What makes it better than VSCode Co-pilot with Claude 4.5? I barely program these days since I switched to PM but I recently started using that and it seems pretty effective… why should I use a fork instead?

Copilot as a harness for the model is generic enough to work with every model. Claude sonnet, haiku, and opus are trained with and specifically for Claude code in mind. Also, as a heavy user of both, there are small paper cuts that seriously add up with copilot. Things that are missing like sub agents. The options and requests for feedback that cc can give (interactive picker style instead of prompt based). Most anno…

I believe you that Claude Code is a better harness, but I'm skeptical it's worth learning because I'm certain that VSCode (Microsoft) will catch up to it eventually. There's nothing differentiated about it, and VSC has already closed the gap. As much as I dislike encouraging BigTech hegemony, it seems obvious which horse to bet will win this race...

Re: Claude Code can debug low-level cryptography

#109

Post by chatmasta »

timr wrote: post #91
chatmasta wrote: post #72

Earlier quoted context omitted.

What makes it better than VSCode Co-pilot with Claude 4.5? I barely program these days since I switched to PM but I recently started using that and it seems pretty effective… why should I use a fork instead?

There’s really no functional difference. The VSC agent mode can do everything you want an agent to do, and you can use Claude if you like. If you want to use the CLI instead, you can use Claude Code (or the GitHub one, or Codex, or Aider, or…) I suspect that a lot of the “try using Claude code” feedback is just another version of “you’re holding it wrong” by people who have never tried VSC (parent is not in this grou…

I think a lot of that feedback is simply an element of how fast the space is moving, and people forming their impressions at different stages of the race. VSCode Copilot today is a wholly different experience than when it first launched as an advanced auto-complete.

Re: Claude Code can debug low-level cryptography

#110

Post by MattGaiser »

rtpg wrote: post #61
simonw wrote: post #11

Using coding agents to track down the root cause of bugs like this works really well: > Three out of three one-shot debugging hits with no help is extremely impressive. Importantly, there is no need to trust the LLM or review its output when its job is just saving me an hour or two by telling me where the bug is, for me to reason about it and fix it. The approach described here could also be a good way for LLM-skepti…

I understand the pitch here ("it finds bugs! it's basically all upside because worst case there's no output anyways"), but I'm finding some of these agents to be ... uhhh... kind of agressive at trying to find the solution and end up missing the forest for the trees. And there's some "oh you should fix this" stuff which, while sometimes isn't _wrong_, is completely besides the point. The end result being these robots…

Just ask it to prioritize the top ones for your review. Yes, they can bikeshed, but because they don’t have egos, they don’t stick to it.

Alternatively, if it is in an area with good test coverage, let it go fix the minor stuff.

Post reply on HN