Hacker Newsnew | past | comments | ask | show | jobs | submit | philips's commentslogin

I have been working on putting agents into full gVisor isolation with a default deny egress proxy that includes secret injections. When an agent tries to connect outbound, by default it suspends the gVisor, holds the proxy connection open, and lets the user choose.

I am adding external judges now. But, there is a hypervisor key you press to deny/allow hosts without switching terminals.

I just can't get behind relying on the harness tool boundry to protect me.


Are putting the agent core and the workspace in separate containers? I do a limited version of that to prevent runaway tests etc from stopping jobs but making it a stronger boundary seems like a good idea.

Yes. Very alpha. https://hshrun.org

Would love to collaborate with you on this since we’re building something analogous internally.

Yes, and all of that changed in 2024 so fairly recently: http://www.kroah.com/log/blog/2024/02/13/linux-is-a-cna/


I think a tricky thing about these metrics is GitHub and the Linux Kernel became their own CNA in mid-2024[1] and mid-2025[2]. This caused a bunch of new automation and lowered barriers to actually filing a CVE.

As a small data point I have a >20k downloads Obsidian plugin and it got a CVE on a Beta release due to a bad pathing regex. I feel this would have never happened 2 years ago.

So, I feel the bar for a CVE being issued is WAYYYY lower in 2025+ than before.

[1]: http://www.kroah.com/log/blog/2024/02/13/linux-is-a-cna/

[2]: https://github.blog/security/vulnerability-research/how-to-r...


GitHub's been doing CVE requests since at least 2022

https://github.blog/security/vulnerability-research/removing...



This is a good point thanks for bringing it up. Will look to capture this nuance in some updates I'm making to the site


If the author is reading the sidebar covers all content on my iPhone 13


There is a button along the headerbar that can be used to toggle the sidebar. One of these days I'll get around to making it actually swipe.


Same, just scroll and it dismiss it.


Not a sophisticated use case but I built a fun service to send handwritten notes to my Pi agent from my eink tablet. It leads to a very satisfying and "slow" workflow.

https://ifup.org/blog/handwritten-notes-with-friends-and-mac...


You are going to revolutionize my life. I was trying to think of cool things to build for my tablet as well as now aiming for specific quirks of certain models is no longer an obstacle and their older custom android forks.

I'll be sending architecture diagrams to my agent, I draw those much faster than I talk. I hope there's a good way to get alerted for agent waiting for your turn. Hmm I wonder if the Claude Code app will install on my Boox. I've been finding it fun to run Claude on weird things, if it runs linux there's a good chance. My favorite install was into my Steamdeck to debug some button sensitivity issues.


Ha! I wrote the first commit to CoreOS's toolbox in 2014: https://github.com/coreos/toolbox/commit/dc4aa0b14bceafd6afa...

And then I read your comment and my first reaction was, oh, they typoed toolbox. But, nope, Red Hat created toolbx which confusingly has the CLI binary of `toolbox`: https://containertoolbx.org


Hi Brandon!


Oh, hey Brian! I like the project. How is it related to reproducible builds though?


... it's not. Just another way to keep my host clean and do the dirty work somewhere else. Where did reproducible builds come through? Another comment?


I just had never heard of "frostyard" and the github.com/frostyard page says: Foundation for Reproducible OS Technologies


I think with things like this. https://github.com/gotgenes/pi-anthropic-auth

I haven’t tested it though


This is exactly why I love e-books as an adult.

As a kid I loved "pocket books"[1] because they didn't make my hand feel weak trying to hold them on a road trip and I could fit a few in my backpack.

But, obviously, not every work was published in a pocket book format.

[1]: https://en.wikipedia.org/wiki/Pocket_Books


These are more generically known as ‘mass market paperbacks’ - and are sadly basically extinct:

https://www.yahoo.com/lifestyle/articles/end-mass-market-pap...

Last year, ReaderLink, the country's largest distributor of books to airports, pharmacies and big-box stores, announced that it would no longer carry mass markets. The books can still be found in some places, but "as a format, I would say it's pretty much over," Ivan Held, the president of publishing imprints Putnam, Dutton and Berkley, said to the Times.

I too mourn them - much more of a convenient size than the now ubiquitous ‘trade paperback’ - but we’re obviously in a small minority.


What are some of your favorite devices/software that follow this idea? I can list mine:

- Supernote by Ratta https://supernote.com

- Air Lab by Networked Artifacts https://networkedartifacts.com

- Xteink X4 (or equivalent) https://www.xteink.com

- Zeroink Writer - my kids use it to practice typing and write letters I can print out. Essentially digital typewriter. https://www.crowdsupply.com/zerowriter/zerowriter-ink

- Readwise - enables me to review highlighted passages and ideas from things I read

- Obsidian - I dump text files into it, it syncs it around for me to review or edit.


I love and use a few of those products and would agree, but I think a bigger component is our relationship to technology.

IMHO, calm technology is both a property of a product and a way of relating to a product. And I think it's quite personal. For example, digital note-taking tablets are anything but calm for me (but I recognize they may be for others!). OTOH, I practically live in Obsidian.

I ask myself:

- Can I have a calm relationship with this product?

- Does the product respect my agency? Can I decide when and how this technology enters my attention?

- Is it configurable? Can I turn off notifications, feeds, recommendations, badges, autoplay, algorithmic prompts, and so one, without crippling the useful parts?

- Is it intentional? Can I open it to get something done and then leave, rather than being pulled into continued use?

- Does it improve my well-being? Am I generally better off? Am I more informed, capable, connected, productive, or am I merely more stimulated (eg. dopamine)?

Some other perhaps non-sexy products: Amazon Echo Dot, my various air sensors, my more reliable home automation components that are already setup and just work, my Kindle and Xteink.


Yeah, calm home automation stuff is the best.

Water main valve robot, Flume water usage sensor, and Eve water sensors near supply lines is super reassuring when traveling or out of the house. So is having a electrical panel that I can check if the oven really was off.


To add to this:

- Playdate (https://play.date) - fun little portable console with an rLCD screen

- Daylight (https://daylight.computer) rLCD paper-like tablet

- Envy (https://envynote.app) - Obsidian without the plugins, local only

- dumb phone (https://dumbphone.so) - make your iPhone dumber, on purpose

- xteink x4 pro (https://www.xteink.com/products/xteink-x4-pro-pocket-ereader) - great, small reader

hope people enjoy!


My favorite is a notebook and pen. I tried all ways to write notes digitally and it only adds stress and friction for me. No batteries to charge, no files to backup, no bluetooth to sync.


Ugh. Writing with pens has always been so uncomfortable for me. (Yes, I've tried good quality pens and different writing techniques.) Typing on a keyboard is so much faster and easier.


Same. I usually sketch diagrams and jot down quick pieces of info. Everything else goes into Obsidian normally.

Though its kind of funny, with my frequent use of agents, my communication skills and ability to articulate ideas has improved quite a bit, so now I use voice dictation a lot more and it's beginning to be faster than keyboard.


Same. While recognizing this doesn’t necessarily hold true for others.


Wow! Good on you for finding the rights and republishing! I would love to read more about the process.


There wasn't much to it --- I contacted the Univ. of Missouri Press, they set a price in terms of initial fee and on-going royalty payments, I returned a signed copy of the contract and they sent a DVD w/ the files.

The only wrinkle was I failed to communicate clearly that I wanted to do a hardcover reprint and they assumed I was doing a softcover (apparently rights are different?) but they decided that the modest quantity I wanted (100) was okay for hardcover (within the variance of unprinted copies from first run?) and allowed it.

I contacted a (sort of) local printer with a digital press to send the files, proofed them, then once approved and printed, I drove in my truck and picked them up (should have parked closer, a couple of boxes took a tumble), stored them in my entry closet and sold them over the course of a couple of years, sending a statement and check each year until the last copy I was selling was sold and I closed the books.

The notable afterthought was that I should have bought used copies at the price that they were then set at by my re-issue and re-sold them after, but I couldn't bring myself to be quite that mercenary.


Thank you for writing up your story. Could you break down the costs at each step? Did you break even on royalties and printing costs?

I am always fascinated by the whole tension of copyright, publishing costs, and the lost knowledge due to the gap between the two.


From uncertain organic memory royalty was a set percentage (7%? came out to a $1,000 check plus a fee $500?).

Print $3,500 as two checks.

Books sold for $70 ea. and if sold on Amazon had $12.83 for their fees.

Sold a few in person, handed out too many for marketing, and did okay on shipping 'cause I re-used packaging. In the end, came out ahead, and would be willing to revisit this sort of thing in the future.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: