Sandboxing AI Agents in Linux(blog.senko.net)
106 points by speckx 19 hours ago | 64 comments
- bigwheels 15 hours agoI use Leash [1] [2] for sandboxing my agents (to great effect!). I've been very happy with it, it provides strict policy-level control for all process-level + network-level activity, as well as full visibility and dynamic runtime controls via WebUI. Way better than bubblewrap imo.
I originally saw it here on HN and have been hooked ever since.
[1] Screenshot: https://camo.githubusercontent.com/99b9e199ffb820c27c4e977f2...
[2] https://github.com/strongdm/leash
Fun fact: Do you know what container / sandboxing system is in most widespread use? Not docker containers, certainly not bubblewrap, and not even full VMs or firecracker. It's Chrome tabs.
[-]- observationist 15 hours agoUsing Chrome for anything seems like a security failure of itself. It's got great features, but damn do they come at a cost.
- necovek 14 hours agoThat's interesting, how does Chrome implement "sandboxing" in Windows and MacOS? For Linux, does it use the same underlying technology as Docker, Podman, LXD, LXC (cgroups, namespaces...)?
Or is a custom "sandboxing" implementation not relying on system level functions (eg. a VM with restricted functions)?
If the latter, I wonder if something like JRE or .NET CLR is still out there in larger numbers, but obviously, Chrome does have billions of users.
[-]- spijdar 13 hours agoYes, Chromium has "native" sandboxing on all those platforms, Windows [0] Linux [1] and MacOS [2].
Chromium uses both seccomp filtering as well as user namespaces (the technology that Docker/Podman use).
The Windows and MacOS sandboxing strategies are more "interesting" because I've seen very few (open source) programs that use those APIs as extensively as Chromium. On Windows, it makes use of AppContainer [3] (among other things), while on MacOS it uses the sparsely documented sandbox API [4], which I think was based on code from TrustedBSD?
[0] https://chromium.googlesource.com/chromium/src/+/HEAD/docs/d...
[1] https://chromium.googlesource.com/chromium/src/+/HEAD/sandbo...
[2] https://www.chromium.org/developers/design-documents/sandbox...
[3] https://learn.microsoft.com/en-us/windows/win32/secauthz/app...
- JCattheATM 12 hours ago> certainly not bubblewrap,
Eh, it might be bubblewrap given it's what flatpak uses.
- kernc 17 hours agoAs a heads up and affirmation that the approach is correct, here's a small shell bubblewrap wrapper that boils the command line down to `sandbox-run claude --dangerously-skip-permissions`.
- sylvinus 15 hours agoThis is the way to go! On my side I've build a very small `claude-vm` wrapper to run each instance in a VM with Lima: https://github.com/sylvinus/agent-vm
- schmuhblaster 4 hours agoMy attempt at a portable solution: Linux VM inside WASM for sandboxed execution: http://agentvm.deepclause.ai
Minimal dependencies, but not as fast as containers or bubblewrap.
- ATechGuy 17 hours agoI will ask what I've asked before: how to know what resources to make available to agents and what policies to enforce? The agent behavior is not predefined; it may need access to a number of files & web domains.
For example, you said: > I don't expose entire /etc, just the bare minimum How is "bare minimum" defined?
> Inspecting the log you can spot which files are needed and bind them as needed. This requires manual inspection.
[-]- senko 16 hours agoArticle author here. I used trial and error - manual inspection it is.
This took me a few minutes but I feel more in control of what's being exposed and how. The AI recommended just exposing the entire /etc for example. It's probably okay in my case, but I wanted to go more precise.
On the network access part, I let it fully loose (no restrictions, it can access anything). I might want to tighten that in the future (or at least disallow 192.168/16 and 10/8), for now I'm not very concerned.
So there's levels of how tight you want to set it.
[-]- ATechGuy 16 hours ago> I feel more in control of what's being exposed and how
Makes complete sense. Thanks for your insights!
- aflag 16 hours agoAsk the agent to bubblewrap itself
- athrowaway3z 17 hours agoI'm launching a SaaS to create yet another solution to the AI Sandboxing problem in linux.
My friends and I have spent a lot of time quietly injecting support down into the kernel without anybody raising a flag, and we finally have the infrastructure in place to solve this problem.
We have also poisoned all the LLMs training data with our approach, so our marketing is primed and we wont even need to learn Claude to use our tool.
We’re planning a soft launch this month, or maybe next month. Depending on how "in the vibe" (our new word for flow :) our team gets.
We’re calling it `useradd`.
Yes, the man page is intimidating, and the documentation is terrible. But once you're over the learning curve, it puts your machine into a kind of 'main frame' mode where multiple 'virtual teletypes' and users can operate on the same machine.
DM me if you want a beta key.
---
Sorry for the snark, but i cringe at the monuments to complexity I see people building, at least this solution is relative simple and free. Still, dont really see what it buys me.
[-]- tasuki 17 hours agoWell done. It took me all the way up to `useradd`...
Edit: too bad about your edit. The comment was just fine without it.
[-]- athrowaway3z 15 hours agoI wrote my comment to vent my disdain for all the circus projects filled with marketing blurbs and features lists for their overengineered vibeslop.
OP is just sharing the cool utility he found, and how it solved a problem for him.
It felt bad to leave them with the message they shouldn't have, or that he's a big part of the problem.
[-]- senko 15 hours agoOP here, no worries, loved the comment and appreciate the feeling :)
- CuriouslyC 15 hours agoI get where this is coming from, and it's not a terrible solution, but VMs are still better in terms of security and isolation. Typical workstation systems are not designed to be secure from their own users, and frontier models are going to get scary good at cracking systems soon.[-]
- carsoon 15 hours agoFully sandboxed VMs are more secure but not everyone is looking for the most secure option. They are looking for the option that works the best for them. I want to be able to share my development environment with the agent, I have a project with 30 1gb and one 30gb sqlite database. I back it up daily and they can all be reconstructed from the code but it takes a long time. When things change I don't want to have to copy them into a separate vm bloating my storage and using excess resources and then having to rectify them, I want to be sharing the same environment with my agent so I can work side-by-side.
I would rather just have the agent not accidentally delete files outside of its working environment but I am not worried about malicious prompt injection or someone stealing my code.
For me I see the LLM as a dumb but positive actor that is trying to do its best but sometimes makes mistakes, so I want to put training wheels on it while still allowing it to share my working space.
- mystifyingpoi 16 hours ago`useradd` doesn't restrict network access.[-]
- kaffekaka 16 hours agoI have used a separate user, but lately I have been using rootless podman containers instead for this reason. But I know too little about container escapes. So I am thinking about a combination.
Would a podman container run by a separate user provide any benefit over the two by themselves?
- eikenberry 12 hours agoWithout any credentials does network access matter?
- senko 16 hours agoI love using different users for separating services I run on the same box!
For development, I want to be able to access/run/modify/delete the files alongside the AI agent. This can be done if groups and group permissions are set correctly (and the agent correctly chmods everything...), but that feels more fiddly than just isolating it with bubblewrap, systemd, or whatever, and preserving the uid/gid.
Just my 2c - it's great that we have options!
[-]- necovek 14 hours agoHey Senko, did you consider using ZFS or BTRFS snapshotting feature to simplify some of the things you need?
For GH auth tokens, you could also pull that outside the sandbox, and have the agent push to a local clone exposed to the host, and local host with no agent automatically push on inotify inside the repo — eg. agent has access to your /agents/scratchpad/my-git-repo, and sync to actual git hosting service like GH (or Launchpad ;) happens with simple script outside it.
- aflag 16 hours agoI don't know if I want to create an ad-hoc list of permissions. What I would like would be something like take a snapshot of my current workspace in a VM. Run claude there and let it go wild. After the end of the session, kill the box. The only downside is potentially syncing the claude sessions/projects. But I don't think that'd be too difficult.[-]
- secure 16 hours agoI recently blogged about how I do this using MicroVMs on NixOS: https://michael.stapelberg.ch/posts/2026-02-01-coding-agent-...
- senko 16 hours ago> take a snapshot of my current workspace in a VM. Run claude there
Sounds like docker + overlayfs might fit the bill, as long as there's a base image that is close enough to what you need.
I don't think there should be One True Way how to run these, everyone can set it up in a way that best fits their workflow.
[-]- ushakov 16 hours agoboth Docker and bubblewrap are not secure sandboxes. the only way to have actually isolated sandboxes is by using VMs
disclaimer: i work on secure sandboxes at E2B
[-]- gf000 14 hours agoWhat about cgroups? I know they are not exactly analogous, but to me that seems like a pretty decent solution.
- senko 16 hours agoNo disagreement from me. From the article:
> Bubblewrap and Docker are not hardened security isolation mechanisms, but that's okay with me.
Edit to add: my understanding is the major flaw in this approach is potential bugs in Linux kernel that would allow sandbox escape. Would appreciate your insight if there are some easier/more probable attack vectors.
- its-summertime 15 hours agoDo you have more information on how to set up such VMs?[-]
- ushakov 15 hours agofor personal use, many ways: Vargant, Docker Sandbox, NixOS VMs, Lima, OrbStack.
if you want multi-tenant: E2B (open-source, self-hosted)
[-]- eikenberry 13 hours agoHashicorp has mostly abandoned Vagrant, so I'd avoid it.
- fsflover 13 hours ago> What I would like would be something like take a snapshot of my current workspace in a VM.
Sounds like you may be interested in Qubes OS, which runs everything in VMs.
- ashishb 13 hours agoI ended up writing my own sandbox so that it works on Mac OS as well and can be used for other tools (but just AI agents) as well[-]
- ATechGuy 13 hours agoCurious to know what made you DIY this?[-]
- ashishb 13 hours agoTell me a better alternative that allows me to run, say, 'markdown lint', an npm package, on the current directory without giving access to the full system on Mac OS?[-]
- ATechGuy 13 hours agosandbox-exec -f curr_dir_access_profile.sb markdownlint[-]
- ashishb 13 hours agoSo you have to install npm package markdownlint on your machine and let it run it's potentially dangerous postinstall step?[-]
- ATechGuy 12 hours agoYou can customize curr_dir_access_profile.sb to block access to network/fs/etc. Why is this not enough?[-]
- ashishb 12 hours agoSome tools do require Internet access.
Further, I don't even want to take the risk of running 'npm install markdownlint' anymore on my machine.
[-]- ATechGuy 8 hours agoI understand the concern. However, you can customize the profile (e.g., allowlist) to only allow network access to required domains. Also, looks like your sandboxing solution is Docker based, which uses VMs on a Mac machine, but will not use VMs on a Linux machine (weak security).[-]
- ashishb 7 hours agoThat's why I wrote my own sandbox. Everyone hand waives these concerns.
Further, I don't know why docker is weak security on Linux. Are you telling me that one can exploit docker?
- waerhert 13 hours agoNice approach! On Ubuntu 24.04 I had to loosen some AppArmor protections by creating a file:
> cat /etc/apparmor.d/bwrap #include <tunables/global> /usr/bin/bwrap flags=(unconfined) { userns, }[-]- amluto 12 hours agoI despise AppArmor and SELinux, especially in cases where they actively get in the way of security like this.
But you shouldn't need to make a global change. Do this:
Or I think you can do this:if [[ -f /proc/$$/attr/exec ]]; then # AppArmor is active. Request "unconfined" for our next exec. echo 'exec unconfined' 2>/dev/null >/proc/$$/attr/exec fi exec ...
(You'd think I'd be more sure of the exact circumstances under which the latter works given that I literally wrote setpriv... At the very least, it will error out if apparmor is not running, which is mildly obnoxious.)$ setpriv --apparmor-profile=unconfined [command]
- enum 14 hours agoI just have an unprivileged secondary local account and do ssh dummy@localhost.
Is this wrong?
- kwar13 13 hours agoI went exactly the same route: https://kaveh.page/blog/claude-code-sandbox
- virtualritz 13 hours agoThis one was posted here recently; works quite well for me:
- HorizonXP 14 hours agoIs this BSD jails' time to shine?
- jhancock 14 hours agoI've started using a container (podman) which is just for the AI tools. I start it up for Codex etc and let it access to the appropriate code directory outside the container.
Anyone else using this approach? Ideas on improvements?
- muggesmuds 16 hours agoWould love this for MacOS[-]
- davidcann 16 hours agoMy app does this on macOS! https://multitui.com
- senko 16 hours agoThere's https://code.claude.com/docs/en/sandboxing that uses something called Seatbelt on Mac and bubblewrap (the same thing I used here) on Linux.
No idea how customizable that is.
- ashishb 13 hours ago
- Jayakumark 16 hours agoSaw something last week using bubblewrap as well in hn github.com/Use-Tusk/fence
- charcircuit 16 hours agoIf you have ssh installed, with network access it can ssh localhost to escape the sandbox.[-]
- qwertox 15 hours agoYou can consider these agents criminals, or treat them like babies. Both can do harm for a while, but one offers a future.
- senko 16 hours agoDon't give it access to your ssh keys![-]
- charcircuit 15 hours agoYes, it should have its own dedicated key instead of sharing one of your own.
- dist-epoch 15 hours ago`ssh localhost` doesn't work for me. maybe because I have enabled only key-based ssh and my user key is not in authorized_keys? am I missing something?[-]
- charcircuit 15 hours agoYou are right in that it would still need to authenticate.
- jauntywundrkind 17 hours agoReally well targeted!
I'd been thinking of using toolbox or devcontainers going forward, but having to craft containers with all my stuff sounds so painful, feels like it would become another full-time job to make containers
Bubblewrap & passing in a bunch of the current system sounds like a great compromise!
I do wonder what isolation something like systemd-run can offer, if that is enough.
Part #2 to me, I also want observability as to what the agent changed. That was one place where containers are such a clear & huge advantage! Having an overlay that contains the changes to the filesystem is so explicit. There's also works like agentfs, that offer a FUSE filesystem backed by Turso DB (sqlite compatible).
[-]- dgl 14 hours ago> Part #2 to me, I also want observability as to what the agent changed.
You could potentially combine https://github.com/binpash/try with bubblewrap (I'm not sure how well they compose and as the docs say it isn't a full sandbox).
The good (and bad because it's confusing and can lead to surprises if misconfigured) thing about Linux containers is all the pieces of containers can be used independently. The "try" tool lets you use the overlay part of containers on your host system, just like Bubblewrap lets you combine the namespacing parts of containers with your host system.
- eikenberry 12 hours agoBubblewrap supports overlayfs mounts [1]. Seems like you should be able to replicate that flow with it.
- longtermop 16 hours ago[dead]
- aktuel 16 hours agoI like this approach for Nix: https://dev.to/andersonjoseph/how-i-run-llm-agents-in-a-secu... It makes it also easy to give the agent only access to the tools it actually needs.