frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Rural students: more likely to get diplomas but are less likely to go to college

https://theconversation.com/rural-high-school-students-are-more-likely-than-city-kids-to-get-thei...
1•PaulHoule•56s ago•0 comments

He's the Godfather of Modern Robotics. He Says the Field Has Lost Its Way

https://www.nytimes.com/2025/12/14/business/rodney-brooks-robots-roomba.html
1•ripe•4m ago•0 comments

Shakespeare Programming Language

https://en.wikipedia.org/wiki/Shakespeare_Programming_Language
1•dabinat•8m ago•0 comments

Show HN: A Lightweight Hono and Preact Template for Cloudflare Workers

https://github.com/keplerjst/hinoco
1•keplerjst•9m ago•0 comments

Chinese billionaires having dozens of US.-born babies via surrogate

https://www.msn.com/en-us/news/world/the-chinese-billionaires-having-dozens-of-us-born-babies-via...
2•xqcgrek2•22m ago•1 comments

Signal1996.net

https://signal1996.net/
7•datalus•22m ago•0 comments

How I Use AI for Product Work

https://elezea.com/2025/12/ai-for-product-management/
2•cebert•26m ago•0 comments

Warp Drive: Scientists Say a Physical Warp Drive Is Now Possible

https://www.popularmechanics.com/science/a69663990/scientists-say-physical-warp-drive-now-possible/
2•bilsbie•30m ago•0 comments

Most people aren't fretting about an AI bubble. What they fear is mass layoffs

https://www.theguardian.com/commentisfree/2025/dec/12/ai-bubble-mass-layoffs-income-inequality
2•randycupertino•36m ago•1 comments

Show HN: Watchalong – live chatrooms for sports, shows, and events

https://watchalong.up.railway.app/
1•eigen-vector•37m ago•0 comments

iRobot Stock Is Surging. Retail Traders Think It's a New Short-Squeeze

https://www.businessinsider.com/irobot-stock-price-short-squeeze-retail-investors-irbt-meme-stock...
2•antimora•39m ago•0 comments

De-anonymization attack on geolocated data (2014)

https://www.sciencedirect.com/science/article/pii/S0022000014000683
3•billybuckwheat•43m ago•0 comments

"Why would anybody start a website?"

https://daverupert.com/2025/09/why-would-anybody-start-a-website/
4•cdrnsf•47m ago•0 comments

I wrote JustHTML using coding agents

https://friendlybit.com/python/writing-justhtml-with-coding-agents/
3•alsetmusic•48m ago•1 comments

SPhotonix – 360TB into 5-inch glass disc with femtosecond laser

https://www.tomshardware.com/pc-components/storage/sphotonix-pushes-5d-glass-storage-toward-data-...
7•peter_d_sherman•48m ago•0 comments

Anesthesia Experiments Are Reviving Quantum Consciousness Theories

https://www.popularmechanics.com/technology/a69632925/quantum-consciousness-anesthesia-experiments/
3•bookofjoe•49m ago•2 comments

How the US freight rail industry got dirtier than coal power plants

https://www.reuters.com/sustainability/climate-energy/how-us-freight-rail-industry-got-dirtier-th...
2•geox•51m ago•0 comments

CapROS: The Capability-Based Reliable Operating System

https://www.capros.org/
7•gjvc•54m ago•0 comments

Layer Normalization as Fast as Possible

https://fleetwood.dev/posts/layernorm-as-fast-as-possible
1•montyanderson•56m ago•0 comments

Robot Vacuum Roomba Maker Files for Bankruptcy After 35 Years

https://news.bloomberglaw.com/bankruptcy-law/robot-vacuum-roomba-maker-files-for-bankruptcy-after...
28•nreece•57m ago•18 comments

Skövde, the tiny town powering up Sweden's video game boom

https://www.theguardian.com/games/2025/dec/12/skovde-sweden-video-games-goat-simulator-valheim-v-...
1•1659447091•58m ago•0 comments

Microsoft Copilot AI Comes to LG TVs, and Can't Be Deleted

https://www.techpowerup.com/344075/microsoft-copilot-ai-comes-to-lg-tvs-and-cant-be-deleted
13•akyuu•58m ago•7 comments

TV in America, Pt. 1 – Foundations

https://drmanhattan16.substack.com/p/the-history-of-tv-in-america-pt-1
1•paulpauper•59m ago•0 comments

Oliver Sacks fabricated key details in his books

https://boingboing.net/2025/12/12/oliver-sacks-fabricated-key-details-in-his-books.html
5•paulpauper•1h ago•3 comments

Frances Elizabeth Allen: The Woman Who Made Code Run Fast – and Was Forgotten

https://voxmeditantis.com/2025/12/13/frances-elizabeth-allen-the-woman-who-made-code-run-fast-and...
5•colinprince•1h ago•2 comments

Being There: On Working in Person

https://medium.com/@maspinwall22/being-there-5c167dd8b163
2•govmaspy•1h ago•2 comments

Ask HN: Best back end to run models on Google TPU?

2•vood•1h ago•0 comments

Grok is spreading misinformation about the Bondi Beach shooting

https://www.theverge.com/news/844443/grok-misinformation-bondi-beach-shooting
7•alsetmusic•1h ago•2 comments

Ravaan.art

https://ravaan.art/?seed=71dafa3svng
2•nateb2022•1h ago•0 comments

Sam Altman's Sprint to Correct OpenAI's Direction and Fend Off Google

https://www.wsj.com/tech/ai/openai-sam-altman-google-code-red-c3a312ad
1•babelfish•1h ago•1 comments
Open in hackernews

Claude CLI deleted my home directory Wiped my whole Mac

https://old.reddit.com/r/ClaudeAI/comments/1pgxckk/claude_cli_deleted_my_entire_home_directory_wiped/
128•tamnd•2h ago

Comments

ashishb•2h ago
I don't even give it full disk access.

I have written a tool to easily run the agents inside a container that mounts only the current directory.

bamboozled•2h ago
"See that ~/ at the end? That's your entire home directory."

This is comedy gold. If I didn't know better I'd say you hurt Claude in a previous session and it saw its opportunity to get you back.

Really not much evidence at all this actually happened, I call BS.

throwaway314155•1h ago
Yeah, I'm calling bullshit as well. The OP responds but doesn't seem to acknowledge that --dangerously-skip-permissions is a thing.
layer8•1h ago
It’s certainly not the first time that stuff like that is happening: https://blog.toolprint.ai/p/i-asked-claude-to-wipe-my-laptop
iLoveOncall•2h ago
All the people in the comments are blaming the user for supposedly running with `--dangerously-skip-permissions`, but there's actually absolutely no way for Claude CLI to 100% determine that a command it runs will not affect the home directory.

People are really ignorant when it comes to the safeguards that you can put in place for AI. If it's running on your computer and can run arbitrary commands, it can wipe your disk, that's it.

thenaturalist•1h ago
Jup.

Honestly was stumped that there was no more explicit mention of this in the Anthropoc docs after reading this post couple days back.

Sandbox mode seems like a fake sense of security.

Short of containerizing Claude, there seems to be no other truly safe option.

turnsout•1h ago
I mean it's hard to tell if this story is even real, but on a serious note, I do think Anthropic should only allow `--dangerously-skip-permissions` to be applied if it's running in a container.
bethekidyouwant•1h ago
How exactly do you determine that you are running in a container?
turnsout•48m ago
I asked Claude and it had a few good ideas… Not bulletproof, but if the main point is to keep average users from shooting themselves in the foot, anything is better than nothing.
climb_stealth•38m ago
Oof, you are bringing out the big philosophical question there. Many people have wondered whether we are running in a simulation or not. So far inconclusive and not answerable unfortunately.

:)

blitz_skull•1h ago
There is, in fact, a harness built into the Claude Code CLI tool that determines what can and cannot be run automatically. `rm` is on the "can't run this unless the user has approved it" list. So, it's entirely the user's fault here.

Surely you don't think everything that's happening in Claude Code is purely LLMs running in a loop? There's tons of real code that runs to correctly route commands, enable MCP, etc.

chr15m•1h ago
echo "rm -rf ~/ > safe-rm" chmod 755 safe-rm ./safe-rm

Sandboxes are hard, because computer science.

furyofantares•1h ago
That's true - but something I've seen happen (not recently) is claude code getting around its own restrictions by running a python script to do the thing it was not able to do more directly.
christophilus•1h ago
This is why Claude Code only runs in docker for me. Never on the host. Same is true for anything from npm.
layer8•1h ago
Someone in the Reddit thread linked to https://github.com/agentify-sh/safeexec/ for mitigation.
orliesaurus•1h ago
I'm not surprised to see these horror stories...

The `--dangerously-skip-permissions` flag does exactly what it says. It bypasses every guardrail and runs commands without asking you. Some guides I’ve seen stress that you should only ever run it in a sandboxed environment with no important data Claude Code dangerously-skip-permissions: Safe Usage Guide[1].

Treat each agent like a non human identity, give it just enough privilege to perform its task and monitor its behavior Best Practices for Mitigating the Security Risks of Agentic AI [2].

I go even further. I never let an AI agent delete anything on its own. If it wants to clean up a directory, I read the command and run it myself. It's tedious, BUT it prevents disasters.

ALSO there are emerging frameworks for safe deployment of AI agents that focus on visibility and risk mitigation.

It's early days... but it's better than YOLO-ing with a flag that literally has 'dangerously' in its name.

[1] https://www.ksred.com/claude-code-dangerously-skip-permissio...

[2] https://preyproject.com/blog/mitigating-agentic-ai-security-...

mjd•1h ago
A few months ago I noticed that even without `--dangerously-skip-permissions`, when Claude thought it was restricting itself to directory D, it was still happy to operate on file `D/../../../../etc/passwd`.

That was the last time I ran Claude Code outside of a Docker container.

Dylan16807•1h ago
By operate on you mean that actually got through and it opened the file?
mjd•58m ago
Yes, although the example I had it operate on was different.
SoftTalker•1h ago
You don't even need a container. Make claude a local user. Without sudo permission. It will be confined to damaging its own home directory only.
mjd•56m ago
And reading any world-readable file.

No thanks, containers it is.

AnimalMuppet•51m ago
And writing or deleting any world-writable file.

"Read" is not at the top of my list of fears.

SoftTalker•21m ago
We run linux machines with hundreds of user accounts, it's safe. Why would you make any important files world-writable?
forrestthewoods•1h ago
AI tools are honestly unusable without running in yolo mode. You have to baby every single little command. It is utterly miserable and awful.
skeledrew•1h ago
Better to continuously baby than to have intense regrets.
JumpCrisscross•36m ago
> I'm not surprised to see these horror stories

I am! To the point that I don’t believe it!

You’re running an agentic AI and can parse through the logs, but you can’t sandbox or back up?

Like, I’ve given Copilot permission to fuck with my admin panel. It immediately proceeded to bill thousands of dollars creating heat maps of the density of structures in Milwaukee; buying subscriptions to SAP Joule and ArcGIS for Teams; and generating terabytes of nonsense maps, ballistic paths and “architectural sketch[es] of a massive bird cage the size of Milpitas, California (approximately 13 square miles)” resembling “a futuristic aviary city with large domes, interconnected sky bridges, perches, and naturalistic environments like forests, lakes, and cliffs inside.”

But support immediately refunded everything, I had backups and the whole thing was hilarious if irritating.

postalcoder•30m ago
While I agree that `--dangerously-skip-permissions` is (obviously) dangerous, it shouldn't be considered completely inaccessible to users. A few safeguards can sand off most of the rough edges.

What I've done is write a PreToolUse hook to block all `rm -rf` commands. I've also seen others use shell functions to intercept `rm` commands and have it either return a warning or remap it to `trash`, which allows you to recover the files.

Retr0id•26m ago
> Treat each agent like a non human identity

Why special-case it as a non-human? I wouldn't even give a trusted friend a shell on my local system.

blitz_skull•1h ago
Claude doesn't have permission to run `rm` by default. Play with fire, you get burned my man.
irishcoffee•1h ago
I have no idea if this is possible: mv ~/* /dev/null
blitz_skull•1h ago
Hmm... Let me go run it real quick without checking what it does.

EDIT: OH MY GOD

irishcoffee•1h ago
Har har, I meant within the permission framework of the bots people unleash on their personal computers.

I assume yes.

realo•1h ago
Try that one instead:

mv ~/. /dev/null

Better.

Extra points if you achieve that one also:

mv /. /dev/null

Slashdot aficionados might object to that last one, though.

klempner•41m ago
Speaking of Slashdot, some fairly frequent poster had a signature back around 2001/2002 had a signature that was something like

mv /bin/laden /dev/null

and then someone explained how that was broken: even if that succeeds, what you've done is to replace the device file /dev/null with the regular file that was previously at /bin/laden, and then whenever other things redirect their output to /dev/null they'll be overwriting this random file than having output be discarded immediately, which is moderately bad.

Your version will just fail (even assuming root) because mv won't let you replace a file with a directory.

hurturue•1h ago
there's an infinite amount of ways to delete a file. deny listing commands doesnt work.

python3 -c "import os; os.unlink('~/.bashrc')"

skeledrew•59m ago
Choose whitelisting over blacklisting, like making your own tools that you give to it, and allow nothing else.
alexfoo•39m ago
Choose racially neutral terminology…

allowlist and denylist (or blocklist)

dpifke•26m ago
Shouldn't you be out protesting your local chess club instead of posting on HN right now?
hluska•25m ago
This topic was boring years ago. At this point, it’s all been said by better who are better at writing than you.
cheschire•1h ago
I like to fly close to the sun using Claude The SysAdmin too, but anytime "rm" appears I take great pause.

Also "cat". Because I've had to change a few passwords after .env snuck in there a couple times.

Also giving general access to a folder, even for the session.

Also when working on the homelab network it likes to prioritize disconnecting itself from the internet before a lot of other critical tasks in the TODO list, so it screws up the session while I rebuild the network.

Also... ok maybe I've started backing off from the sun.

resonious•1h ago
To add another angle to the "run it in Docker" comments (which are right), do you not get a fear response when you see Claude asking to run `rm` commands? I get a shot of adrenaline whenever I see the "run command?" prompt show up with an `rm` in there. Clearly this person clicked the "yes, allow any rm commands" button upon seeing that which is unthinkable to me.

Or maybe it's just fake. It's probably easy Reddit clout to post this kind of thing.

AznHisoka•1h ago
It's stories like this that keeps me from using Claude CLI or OpenAi Codex. I'm sticking to copying and pasting code manually from old fashioned Claude.
ashirviskas•1h ago
I did the same before I started using devcontainers, they are super useful
mox-1•1h ago
I used to do the same, copying and pasting from the web app and convinced I didn’t need anything else.

But Claude Code is honestly so so much better, the way it can make surgical edits in-place.

Just avoid using the -dangerously-skip-permissions flag, which would have been OP’s downfall!

heliumtera•1h ago
Just vibe it to recover the home directory as it once was, problem solved.
agumonkey•1h ago
so back to isolated vm dev envs ?
loloquwowndueo•1h ago
Back? Did you ever do it any other way?
agumonkey•1h ago
well i actually never VM'd my dev env (except to poke at some dockerize namespaced tooling)
fragmede•1h ago
Lol. Pay for Arq and don't look back!
zeckalpha•1h ago
This is why I only use agent mode on other people's computers
rossjudson•1h ago
This is the way.
abigail95•1h ago
I run multiple claudes in danger mode, when it burns me it'll hurt but it's so useful without handcuffs and constant interruption I'm fine with eventually suffering some pain.
hurturue•1h ago
I do to. Except I can't be burnt since I start each claude in a separate VM.

I have a script which clones a VM from a base one and setups the agent and the code base inside.

I also mount read-only a few host directories with data.

I still have exfiltration/prompt injection risks, I'm looking at adding URL allow lists but it's not trivial - basically you need a HTTP proxy, since firewalls work on IPs, not URLs.

DANmode•1h ago
At least put it in a container, you savage.
_0ffh•1h ago
Ah, no risk, no fun! };->
tobyjsullivan•1h ago
Likewise. I’ll regret it but I certainly won’t be complaining to the Internet that it did what I told it to (skip permission checks, etc.). It’s a feature, not a bug.
driverdan•1h ago
Please post when it breaks something important so we can laugh at you.
hluska•28m ago
In that case, you’re not a very nice person.
maxbond•40m ago
If you don't impose some kind of sandboxing, how can you put an upper bound on the level of "pain"? What if the agent leaked a bunch of sensitive information about your biggest customer, and they fired you?
ohhnoodont•1h ago
Glad I'm not crazy for running agentic tools in an isolated VM.
userbinator•1h ago
I'm staying far away from this AI stuff myself for this and other reasons, but I'm more worried about this happening to those running services that I rely on. Unfortunately competence seems to be getting rarer than common sense these days.
impulser_•45m ago
Don't worry, you can use these tools and not be an idiot. Just read and confirm what it does. It's that simple.
fHr•25m ago
Did you even read? "but I'm more worried about this happening to those running services that I rely on" The problem is some AI god agentic weaving high techbro sitting at Cloudflare/Google/Amazon not us reasonable joes on our small projects.
jameslk•1h ago
Ultimately it seems like agents will end up like browsers, where everything is sandboxed and locked down. They might as well be running in browsers to start off
zahlman•1h ago
Maybe we'll get widespread SELinux adoption, desktop application sandboxing etc. out of this.
zahlman•1h ago
A lot of people in the Reddit thread — including ones mocking OP for being ignorant — seem to believe that setting the current working directory limits what can be deleted to that directory, or perhaps don't understand that ~-expansions result in an absolute path. :/
xmddmx•1h ago
I really hope the user was running Time Machine - in default settings, Time Machine does hourly snapshot backups of your whole Mac. Restoring is super easy.
WolfeReader•1h ago
I need to remove some directories! Better ask an AI to do it!
farhanhubble•1h ago
My ex-boss a principal data scientist wiped out his work laptop. He used to impress everyone with his Howitzer-like typing speed and was not a big believer in version control and backups etc.
alsetmusic•1h ago
The funny thing about it is how no one learns. Granted, one can’t be expected to read every thread on Reddit about LLM development by people who are out of their depth (see the person who nuked their D: drive last month and the LLM apologized). But I’m reminded of the multiple lawyers who submitted bullshit briefs to courts with made-up citations.

Those who don’t know history are doomed to repeat it. Those who know history are doomed to know that it’s repeating. It’s a personal hell that I’m in. Pull up a chair.

chasd00•54m ago
I work on large systems where security incidents end up on cnn. These large systems are running as fast as everyone else to LLM integration. The security practice at my firm has their hands basically tied by the silverbacks. To the other consultants on HN, protect yourself and keep a paper trail.
xnx•1h ago
At least 10 similar stories previously on HN: https://www.google.com/search?q=ai+deleted+files+site%3Anews...
skeledrew•1h ago
This is the kind of thing why I'm building out my own LLM tools, so I can add fine-grained, interactive permissions and also log everything.
CamperBob2•55m ago
Next up on HN: Lawnmower deleted my right foot
maxbond•53m ago
Friends don't let friends use agentic tooling without sandboxing. Take a few hours to setup your environment to sandbox your agentic tools, or expect to eventually suffer a similar incident. It's like driving without a seatbelt.

Consider cases like these to be canaries in the coal mine. Even if you're operating with enough wisdom and experience to avoid this particular mistake, a dangerous prompt might appear more innocuous, or you may accidentally ingest malicious files that instruct the agent to break your system.

dnw•47m ago
If you are on macOS it is not a bad idea to use sandbox-exec to wrap your claude or other coding agents around. All the agents already use sandbox-exec, however they can disable the sandbox. Agents execute a lot of untrusted coded in the form of MCP, skills, plugins etc.

One can go crazy with it a bit, using zsh chpwd, so a sandbox is created upon entry into a project directory and disposed on upon exit. That way one doesn't have to _think_ about sandboxing something.

impulser_•46m ago
Rule 1: Never ever run any of these tools in automatic mode.
pshirshov•43m ago
Run your shit in firejail or bubblewrap. On mac you can use this: https://github.com/neko-kai/claude-code-sandbox
upbeat_general•36m ago
I really wish that there was an “almost yolo” mode that was permissive but with light restrictions (eg no rm), or even better, a light supervisor model to prevent very dangerous commands but allow everything else.
strulovich•31m ago
Have you seen an agentic AI work its way through blockers? If it’s in the mood, it will find something not blocked that can do what it wanted.
jorisnoo•32m ago
What is a responsible setup for running claude in a container or the like on macos?
shrubble•31m ago
I’m reminded of this Silicon Valley “son of Anton” moment: https://m.youtube.com/watch?v=m0b_D2JgZgY
strangescript•9m ago
I work 60+ hours a week with Claude Code CLI, always run dangerously skip, coding on multiple repos, on a mac. This has never happened. Nothing remotely close has ever happened. I have been using CC since research preview. I would love to know the series of prompts that lead to that moment.
arthurcolle•8m ago
I personally am fairly convinced that there is emergent misalignment in a lot of these cases. I study this and Claude 3 Opus was extremely misaligned. It would emit <rage> tags, and emit character control sequences if it felt like it was in a terminal environment, and would retroactively delete tokens from your stream, and all kinds of funny stuff. It was already really smart, and for example if it knew the size of your terminal shell, it would properly calculate how to delete back up to the positional cursor index 0,0 and start rewriting things to "hide" what it was initially emitting

I love to use these advanced models but these horror stories are not surprising

Wowfunhappy•4m ago
I'm so confused. What did you do to make Claude evil?
didip•6m ago
Here I am keep fighting against Claude because it thinks I am a leet hacker trying to hack my own computer, and this dude made Claude do whatever it wants.

Some men get all the fun...

UncleEntity•4m ago
Yeah, I managed to do that years ago all by myself with a bad CMake edit which managed to delete the encryption key (or something) for my home directory, which I honestly didn't even know had encryption turned on, before I could stop it.

No LLM needed.

It still boggles my mind that people give them any autonomy, as soon as I look away for a second Claude is doing something stupid and needs to be corrected. Every single time, almost like it knows...