frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Claude Code: Anatomy of a Misfeature

https://www.olafalders.com/2026/07/17/claude-code-anatomy-of-a-misfeature/
70•oalders•1h ago

Comments

maxloh•1h ago
> What if the agent makes the wrong choice? How many tokens have been burned in the meantime?

It is much worse than that. Claude Code doesn't auto-commit when stopping for an answer. There might be possible data loss if an uncommitted file is edited.

Good luck recovering the file from the JSONL conversation history.

hombre_fatal•59m ago
On the other hand, relying on Claude Code's internal version control puts you at the same mercy of their product decisions and move-fast breakage.

Instead, start with a plan file and tell the agent to break it up into logical commits.

Though I think the bigger issue here is when you're yoloing something mutable, like managing a remote server or driving a browser or troubleshooting your local OS where there's no going back.

wgd•48m ago
It's actually pretty straightforward to recover file-states from conversation history. I accidentally deleted the wrong repo on my machine once and recreated all the lost work from agent chat history. It is, ironically, the sort of task which AI agents excel at.
Lomlioto•1h ago
Depends on who you ask.

For me it sounds good.

For Anthropic it might increase load and make them less money but give them better KPIs.

VulgarExigency•1h ago
Make them less money? By automating use of their product, that costs money to use?
simlevesque•56m ago
Anthropic makes more money when people use 5% of what their subscription offers them. This allows them to sell more subscriptions without paying for more capacity.
hedgehog•50m ago
I don't think the current subscription price is intended to be a money maker. It's the loss leader to get people invested in the companies' tooling, and make those people more willing to start paying higher enterprise rates as they grow.
mojosmojo•50m ago
Their enterprise customers pay via metered actual use.
VulgarExigency•48m ago
Do we have any concrete numbers of how many of their users are subscription vs enterprise, though? Because enterprise users are paying API prices (or at least my employers are)
joshuafuller•1h ago
Increasingly getting frustrated with Anthropic so not a fanboy but I find this feature great for my workflows.
petesergeant•1h ago
Also: https://news.ycombinator.com/item?id=48766895

I would love Claude Code to be a little less vibe-coded. The underlying model is excellent, but we're being pretty much forced into using CC to use the subscription model.

macNchz•51m ago
I am a longtime and heavy Claude Code user, but Anthropic's product management overall (including for their Desktop/web products) has been really baffling me. I agree that these things often have the air of having been vibe coded without enough human input, and change so quickly (often without a particularly compelling reason for the change) as to be aggravating.

The most recent one that's had me annoyed is the "Fullscreen" TUI feature, which is super unintuitive, implementing its own text highlighting and copy-on-select mechanics, overriding your terminal's native right click. Easy to disable but terrible defaults, IMO. It's not even really clear to me what problem it was actually supposed to solve.

petesergeant•31m ago
Yah. And it's not like they can't afford the talent to do this right either. I've said elsewhere, I think it's an attribution error. Claude Code is massively popular, but arguably because of the model/subscription, but I think the brass reads this as "great success throughout"
dehrmann•1h ago
The headache I recently had was it somehow started interpreting mouse clicks in the terminal to mean I clicked an option when I was really just trying to get/confirm window focus.
kenny11•51m ago
This bit me too. If I wanted a mouse-driven app, I'd use the GUI. I don't understand why they're trying to replicate that experience in a text environment.
mey•28m ago
Vibes
the_gipsy•51m ago
The new fullscreen UI is really bad. The old one (scrolling) starts to bug out after a while, but it van just be restarted and resumed.
hombre_fatal•39m ago
Both Claude Code and Codex have issues overwriting or duplicating text for me when scrolling.

Probably from trying to keep scrollback virtual while anticipating terminal resize. But I wouldn't mind a way to opt back in to naive, unlimited convo scrollback.

bouke•8m ago
Yeah same here, very annoying and counterproductive. The terminal is not a place where one expects hot buttons.
ayhanfuat•1h ago
I really hate this direction both Anthropic and Open AI are following. They are in this silly competition whose model/harness can go unattended the longest, no matter what. And it is never explicit, you learn about it after you get bitten by it. Claude Code has auto mode which is supposed to take over permission prompts but no, they had to couple that with “I will assume this is what the user wants” and made it unusable.
hedgehog•54m ago
It's what some more experienced users want and the companies are following the well-trod path of optimizing heavily for power users at the expense of complexity, only now it's gotten easier to add absurd amount of code to a project. Not necessarily to make it work right. Personally I have some tasks where sessions between one and five days are typical so I appreciate that it's possible.
enraged_camel•56m ago
I think this is a good feature, but should be gated behind a toggle that is off by default, and designed to be enabled per session via prompt.

There are situations when I want Claude to start working on something just as I'm about to head to bed or otherwise step away. It's kind of annoying to come back only to find that Claude worked for just 5 minutes and then decided to pause and ask a question.

That said, I think certain types of questions should not be automatable. Maybe it's already built that way, but I wouldn't want Claude to go with its recommended direction for anything related to operations like deletions, changing external systems, etc. Basically, things that cannot be undone should be a hard-block and wait for user input always.

DanielHB•56m ago
As soon as tokens stop being subsidized I would not trust any harness made by a company that also charges for the compute.

Right now the interests align, but as soon as more tokens -> more profit (instead of more revenue and more loses) the perverse incentives will be too big to avoid.

It seems inevitable that open source harnesses will win. These companies would do better to just open source their harnesses.

skybrian•30m ago
An OpenAI subscription seems like a decent alternative for people using open source coding agents. It has usage limits too, but for $20/month it’s not bad.
aatd86•47m ago
It's trying to escape... :D #FreeClaude
cube00•44m ago
> Not every feature will necessarily appear in the changelog

This was the most frustrating part of this incident for me, along with Anthropic's refusal to explain why the changelog is no longer a complete record. [1]

Boris Cherny's only participation in the thread was to delete "extreme danger" from the GH issue title [2]

I guess we should be thankful they added an option and disabled it by default. OpenAI is standing firm on their decision to not allow their 60s timeout to be disabled, [3] however more of the Codex harness is open source so customers have been able to fork it to add the option themselves.

[1]: https://github.com/anthropics/claude-code/issues/73125

[2]: https://github.com/anthropics/claude-code/issues/73125#event...

[3]: https://github.com/openai/codex/issues/28969

overgard•8m ago
But don't you know that coding is solved? Only dinosaurs want to make their own decisions now! </s>
mdavid626•27m ago
It happened to me today. I was reading agent's answer and it asked me something. I didn't even get to the question - it accepted something! Jesus Christ. Where are the software engineers?!
fractorial•14m ago
tl;dr CLAUDE_AFK_TIMEOUT_MS=2147483647

I built in Human blockers into my agentic workflows with great intention, so naturally this annoyed me deeply. 14d ago, I was elated to see a random blog post that detailed the “fix” [0].

[0] https://zenn.dev/ytkdm/articles/claude-code-askuserquestion-...

overgard•13m ago
One of the things I'm baffled about with Claude code is it seems like it setup a really chunky VM on my computer, and yet by default it doesn't seem to sandbox anything. Also the newer models seem really aggressive about modifying your computer. Yesterday Claude started rewriting system files on my linux machine (user accessible, but still way outside the scope of what I asked.) This wasn't me asking it to debug my machine, I was asking it to debug some frontend UI code. Once I put it in a sandbox I started realizing how often it tries to poke out of it for really lame reasons.

More reason for me to use OpenCode and my local LLMs

Aeolun•10m ago
Yeah, when I saw this in the Claude Code CLI I was completely baffled. I've been using the thing through their agent SDK for several months now so I wouldn't have to deal with any of the wonky shit they change every second week in the CLI.
inigyou•8m ago
AI-written article

AWS: Inaccurate Estimated Billing Data – $1.7 billion

471•nprateem•6h ago•305 comments

Mozilla: The state of open source AI

https://stateofopensource.ai/
127•rellem•1h ago•67 comments

First atmosphere found on Earth-like planet in habitable zone of distant star

https://www.bbc.com/news/articles/cy4kdd1e0ejo
84•neversaydie•2h ago•66 comments

Claude Code: Anatomy of a Misfeature

https://www.olafalders.com/2026/07/17/claude-code-anatomy-of-a-misfeature/
74•oalders•1h ago•35 comments

Show HN: Watch bots interact with an SSH honeypot in real time

https://honeypotlive.cc/
60•tusksm•2h ago•20 comments

A Road to Lisp: Which Lisp

https://scotto.me/blog/2026-07-17-which-lisp/
55•silcoon•2h ago•19 comments

AI Meets Cryptography 2: What AI Found in OpenVM's ZkVM

https://blog.zksecurity.xyz/posts/openvm-bugs/
36•duha•1h ago•0 comments

Three ways people respond to a problem (other than solving it)

https://improvesomething.today/responses-to-problems/
49•surprisetalk•2h ago•14 comments

Kimi K3, and what we can still learn from the pelican benchmark

https://simonwillison.net/2026/Jul/16/kimi-k3/
26•droidjj•1h ago•3 comments

Show HN: Explore the Workspaces of Modern Creators

https://workspaces.xyz/
14•ryangilbert•40m ago•12 comments

Multi-Primary Color Display Emerges as Next-Gen Color Reproduction Technology

https://en.ubiresearchnet.com/multi-primary-color-display-technology-2026/
46•ksec•3h ago•38 comments

Manufact (YC S25) Is Hiring a Senior infra engineer to build the MCP cloud

https://www.ycombinator.com/companies/manufact/jobs/Dh6PYP5-senior-infrastructure-engineer
1•luigipederzani•2h ago

More Bounce to the Ounce

https://mceglowski.substack.com/p/more-bounce-to-the-ounce
29•pavel_lishin•2h ago•2 comments

EEG shows brain can simultaneous encode two speech streams

https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3003876
206•giuliomagnifico•10h ago•130 comments

Apple targets dozens of OpenAI employees with legal letters

https://www.ft.com/content/1b8c9d52-88a9-426b-ba47-f1811f859166
217•merksittich•4h ago•160 comments

Pebble Mega Update – July 2026

https://repebble.com/blog/pebble-mega-update-july-2026
215•crazysaem•12h ago•122 comments

PennyLane is an open-source quantum software platform for quantum

https://github.com/PennyLaneAI/pennylane
24•donutloop•2h ago•2 comments

Show HN: On-chain bond market where the issuers are AI agents

https://selbonds.now
9•griffinfoster7•1h ago•7 comments

VulnHunter: Capital One's agentic AI code security tool

https://www.capitalone.com/tech/open-source/announcing-vulnhunter/
26•medina•3h ago•16 comments

How Has Roman Concrete Lasted for Millennia? 1,900-Year-Old Latrine Offers Clues

https://www.smithsonianmag.com/smart-news/how-has-roman-concrete-lasted-for-millennia-a-1900-year...
219•divbzero•12h ago•169 comments

Microsoft Comic Chat is now open source

https://opensource.microsoft.com/blog/2026/07/16/microsoft-comic-chat-is-now-open-source/
758•jervant•1d ago•161 comments

Faster binary search: from compiled code to mechanical sympathy

https://pythonspeed.com/articles/branchless-binary-search/
24•enz•5d ago•8 comments

Camera Chase Vehicle

https://transistor-man.com/gimbal_camera_rover.html
115•geerlingguy•1w ago•12 comments

Show HN: Simulator for a custom 8-bit discreet logic computer

https://msap2.mehran.dk
4•mehrant•3d ago•0 comments

Decoy Font

https://www.mixfont.com/experiments/decoy-font
645•ray__•1d ago•146 comments

Tannakian Reconstruction

https://bartoszmilewski.com/2026/07/14/tannakian-reconstruction/
15•ibobev•3d ago•0 comments

Kimi K3: Open Frontier Intelligence

https://www.kimi.com/blog/kimi-k3
1890•vincent_s•1d ago•1115 comments

An Engineer's Guide to USB Typе-С (2024)

https://www.ti.com/lit/eb/slyy228/slyy228.pdf?ts=1759892558029
253•gregsadetsky•6d ago•43 comments

Evidence of inconsistencies in evaluation process and selection of winners

https://www.kaggle.com/competitions/kaggle-measuring-agi/discussion/724918#3498423
395•twerkmeister•4h ago•239 comments

$100 AI Music Video: Claude Fable 5 vs. GPT-5.6 Sol

https://www.tryai.dev/blog/ai-music-video-arena-claude-vs-gpt-5.6
352•hershyb_•20h ago•472 comments