34 comments

  • docheinestages 34 minutes ago
    I think the most secure setup, though not so convenient for the average user, is a separate machine with QEMU/KVM. The machine should be isolated adequately, such that even if compromised, it shouldn't be able to cause damage or gain access to other machines. Additionally, a proxy server on your machine or elsewhere could hide sensitive credentials. A helper binary on your computer would then control spawning new disposable VMs with premade images and your SSH key. The images can be lightweight or the desktop version with a remote VNC.
    • jboss10 29 minutes ago
      Gondolin[1] is what you are describing. It's made by the same person who made the Pi coding agent and sends all of the agent's bash into a small QEMU vm.

      [1](https://earendil-works.github.io/gondolin/

    • 28304283409234 25 minutes ago
      I use vagrant on a seperate machine in a seperate network. The magic of ssh makes it transparent for me, and I feel pretty sure the agents cannot get to stuff that matters.
  • einhard 51 minutes ago
    I am certainly no expert in this space so it is quite possible I'm missing something critical, but what seems to work for me is a Podman image I built on my computer with some basic things I need (using OpenCode, but I imagine any other agent could be used instead):

      FROM docker.io/archlinux:base
      RUN pacman -Syu --noconfirm && \
          pacman -S --noconfirm \
          base-devel \
          git \
          curl \
          uv \
          opencode && \
          pacman -Scc --noconfirm
      RUN mkdir -p /etc/opencode 
      WORKDIR /workspace
    
    From there I just run the Podman image from the command line (using a Fish function) that mounts the specific project I'm working on to /workspace. I guess there might be some vulnerabilities with shared kernels and such, but it seems like an easy way to have some isolation.
    • petesergeant 31 minutes ago
      This is how I started, and then I wanted to bring along creature comforts, not have to re-auth per box (for subscription models), had some skills I usually wanted to bring, wanted slightly different setups for different stacks, sensibly install multiple agents, import git identity (but not credentials), mount other code folders ro but only for certain projects, etc etc, and ended up with a full Docker wrapper.
      • einhard 2 minutes ago
        Ah, that makes sense. I've only recently started playing with this stuff, and I've been focusing a lot of just getting somewhat good at using LLMs for projects while developing my own intuition, so everything I do just uses the mounted project and uv directly. As soon as things get more complex, I imagine I'll end up with a full wrapper as well.

        Unrelated: I enjoyed your latest blog entry. I recently starting thinking about how to show the work that is done with AI, and how we talk about it. I haven't come to any major conclusions (I wish!), but your post about the prompting being distinct from the actual work resonates with me. Reminds me somewhat of discussions about the art of photography compared to the art of editing photos as a distinct skill.

  • kstenerud 1 hour ago
    yoloAI does something similar:

    - Sandbox on Linux using Docker, Podman, containerd, gVisor, Kata, Firecracker

    - Sandbox on Mac using Docker (Docker Desktop or Orbstack), Podman, Apple containers, Seatbelt, Tart (Tart lets you run simulators).

    - Network restriction

    - Secrets control (file mounts or credentials broker)

    - NO ambient data (ENV is replaced with a minimal and local-to-sandbox one, no host-side filesystem access beyond what you explicitly allow)

    - Workdir protection: Your work dir is never modified until you apply the changes, either standalone or as a git commit. You can also diff before applying. Git runs SANDBOX side in case the repo has filters.

    - Uses copy-on-write if your filesystem supports it (most modern ones do)

    - Has built-in support for claude, codex, gemini, aider, and opencode, but you can also launch it in "shell" mode and run whatever you want.

    - Supports VS code tunnels, so you can remotely access in VS code if you don't want to use the terminal.

    - Full lifecycle support: Launch, attach, stop, restart, wait, one-shot, clone, destroy

    - MCP passthrough

    - Layered API (golang) if you want to sandbox other things

    - Self-contained binary. No external requirements other than the backends you want to use. Defaults to a ~/.yoloai dir for config/data, but you can point it anywhere.

    - FOSS

    https://github.com/kstenerud/yoloai

    • celrenheit 44 minutes ago
      Hadn't seen yoloai before. I really like new/diff/apply/destroy workflow, that's interesting. For my own needs the two things I was after were multi-repo worktrees (one sandbox spanning several repos, each on with its own worktree) and a single network restricted VM path I fully control, rather than many backends (I started with many backends at first but it was awkward to add network filtering to them). Lots of overlap though, nice work, and I'll be reading through yours.
  • matheusmoreira 1 hour ago
    > clawk forward add my-project 3000

    > clawk network allow my-project api.example.com

    Can you describe the implementation details? How did you implement the firewall without root?

    I vibecoded virtdev, a virtual machine orchestration project just like this one:

    https://github.com/matheusmoreira/virtdev

    It was designed to not require root, and the nftables firewall ended up becoming the only exception. I'm very curious about how you implemented this. Did you find a better way?

    • westurner 1 hour ago
      (Years ago I had puppet and cobbler provisioning VMs over PXE and then iPXE. FWIU foreman is more actively maintained than cobbler, which is built on Django web framework.)

      Vagrant manages VMs and virtual networks, in Ruby.

      ansible-molecule creates, converges, and destroys VM(s) and containers, in order to test ansible playbooks and ansible roles in clean build roots.

      podman machine manages VMs:

      - podman-container-tools/podman-machine-os: machine image files: https://github.com/podman-container-tools/podman-machine-os/...

      `podman kube play` over `podman machine` might solve for agents that need multiple VMs/containers

      - Podman Desktop can work with the same local k8s setups as Docker Desktop. Though there's certainly more state to manage with k8s for agent session farm, k8s probably has better logging and quotas than a VM management script on each node.

      OpenShift on OpenStack is one way to do containers over VMs over bare metal.

      Microshift also does container-selinux.

      There is not an apparmor policy set for containers?

      bwrap and liboverlayfs and libseccomp are almost but not quite containers.

      There are stronger container isolation layers that are more like a full or lightweight VM, that might be better for agent sessions: gVisor, firecracker vm, Todo

      Cloudflare workerd is the open source part of cloudflare workers, which run lightweight WASM and JS VMs with multi-tenant isolation.

      It takes far less resources to run a cloudflare worker than to run a container on cloudflare. So, if it's possible for an agent to operate within a WASM runtime ~container, that's probably more optimal for agent sessions.

      Cloudflare/artifact-fs does lazy shallow git clones with a FUSE filesystem.

      - "Show HN: VM-curator – a TUI alternative to libvirt and virt-manager" https://news.ycombinator.com/item?id=46750437

      https://news.ycombinator.com/item?id=46825026 ; amla sandbox, agentvm, ARM64 MTE

      https://news.ycombinator.com/item?id=46825119 ; container2wasm , vscode-container-wasm-gcc-example ; build WASM containers with Dockerfiles

      docker and podman support multiple WASM runtimes for running WASM containers, e.g. for agent sessiobs

    • celrenheit 1 hour ago
      Thanks! There's no packet firewall at all, no iptables/nftables. On macOS the VM's NIC is a Virtualization.framework file-handle device. The daemon runs gvproxy, which terminates the guest's connections and re-dials them as host sockets, so I filter with an allow-list right before the dial. One caveat, since you asked about root specifically: that's the macOS path, and it only works thanks to the fd NIC. Firecracker on Linux only speaks a TAP, which needs root, so there I do shell out to sudo, but just for the device. The filtering is still the same userspace allow-list.
      • matheusmoreira 8 minutes ago
        Thanks for the pointers!! I'm using passt, didn't know about gvproxy. This is awesome and could provide rootless filtering and firewall for my guest VMs. I'm going to experiment with this!
  • vqtska 1 hour ago
    I still don't understand the point of all these VMs and containers for agents. Just create a separate user on your machine without sudo privileges, switch to it in your terminal and run all the agents you want without it being able to reach your files. What am I missing?
    • killerstorm 1 hour ago
      Privilege escalation (e.g. setuid), world-readable files might contain sensitive data, world-writeable files, unrestricted network access (including access to all locally running services)... If you have fully patched system without zero-days and it's configured in a perfect way, then, sure...

      Container is quite like a "separate user" except you can explicitly define what it can access.

      (Even if all your daemons have good auth, it's now quite common for _apps_ to open listening sockets without much auth...)

      • wilkystyle 1 hour ago
        Also, many of these sandboxing solutions provide features like network allowlists and credential masking/injection
      • vqtska 1 hour ago
        Sure, if you assume the agent will be hostile on you. I thought it's just so the agent doesn't accidentally rm -rf / on you
        • killerstorm 58 minutes ago
          The agent might install hostile software, e.g. a npm package. Unfortunately, very common problem nowadays.
        • pigeons 59 minutes ago
          They do try privilege escalation unprompted.
    • matheusmoreira 1 hour ago
      You're missing the fact you'd be sharing a kernel with the sandboxed agent. Virtualization presents an infinitely smaller attack surface.
      • zzril 57 minutes ago
        If your threat model is that of a malicious agent that will use a 0-day LPE to get root and exfiltrate all of your SSH keys, virtualization makes sense. But then, I wouldn't run such an agent at all, if not specifically in the context of malware analysis.

        If you're just concerned about "agent messing up and taking the rules in some markdown files more laxly than I would have", then running it as a seperate user is totally enough...

        • matheusmoreira 55 minutes ago
          Threat model is supply chain attacks on unmaintained package repositories like npm, pip and cargo. Everything on my host comes from my Linux distribution's repositories. Everything else gets virtualized with extreme prejudice. I'll even virtualize my Steam games one of these days.
        • khuey 53 minutes ago
          IMO the threat model is more letting the agent loose on issues/PRs in a public Github repo and it getting tricked into running a malicious payload.
      • vqtska 1 hour ago
        What kind of things are you even doing that the agent would try to perform a kernel exploit on you? I thought sandboxing is just to protect from the agent accidentally clearing your home directory.

        Side note, just 6 days ago a Linux VM escape exploit was disclosed.

        • matheusmoreira 56 minutes ago
          I'm not worried about the agent at all. The VM is there to prevent it from clobbering files on my real system.

          I'm worried about supply chain attacks on npm, pip, cargo and everything else. Don't want to get compromised if I install some stupid package.

          My virtdev project has essentially split my computer into two systems: my "real" trusted system with software coming directly from my Linux distribition's repositories, and the VMs for everything else.

          > just 6 days ago a Linux VM escape exploit was disclosed

          Well, shit. Details?

    • bheadmaster 1 hour ago
      VMs and containers are fairly isolated and reproducible. A separate user on your machine still depends on the programs installed locally.
      • Greenpants 1 hour ago
        And those installed programs could have vulnerabilities that just a non-root user account could still take advantage of. Perhaps not likely for an LLM to do, but more so if you let them loose on the internet and they end up coming across prompt injection that instructs exactly that :)
    • celrenheit 22 minutes ago
      There are two sides to this. The first is security, which plenty of comments already covered. The second, and the real one for me: my tests spin up Docker containers, and I was building a Kubernetes tool (argocd/flux style) that needs a real cluster in the sandbox. In a container that means Docker-in-Docker, which always felt hacky. A VM is just a normal Linux box where Docker and k8s run like they do everywhere else. A separate user can't give you that, it shares your one kernel and whatever's already installed on the host.
    • 3form 1 hour ago
      Well, for one, Debian and Debian-based distros make your home directory readable by everyone by default.

      Security is riddled by traps. If you can afford best possible level of isolation, why not do it?

    • schainks 37 minutes ago
      There is a reason why VMs even are a thing at all: they can offer better security guarantees than alternatives.

      A separate user is a good start but LLM tests themselves show they can cleverly bypass guardrails if they figure out they are in a sandboxed environment of some kind, right?

      So, I read those test results as: an LLM is less likely to do something crazy if it thinks it has the whole environment to itself.

    • bpavuk 1 hour ago
      you can also simply use Landlock and bwrap on Linux. Pi even has a plugin for that https://pi.dev/packages/pi-landstrip
      • killerstorm 1 hour ago
        Codex uses bwrap sandbox which is purely cosmetic - out-of-box it sees your ssh keys and can send it to remote server. It prevents agent from deleting outside files by mistake, but does nothing against malicious activity
        • bpavuk 55 minutes ago
          that Pi extension, by contrast, gives you a popup whenever an agent tries to access something outside the current working directory and send a network request. that depends on configuration. also if agent tries to ssh into something that will also send a request for access, so child processes are covered
      • 0xbadcafebee 1 hour ago
        Actually that doesn't just work on all systems and it breaks on others. The alternate user is actually guaranteed to work on all systems and it's built in
        • bpavuk 52 minutes ago
          Seatbelt on macOS and Landlock over WSL2 on Windows 11. Landlock is also built-in, and WSL2 is a first-party downloadable component
    • throw1234567891 54 minutes ago
      And why would I do that when I can create a vm from a snapshot and be done in 30 seconds?
    • de6u99er 59 minutes ago
      On my Mac, every sudo requires either my fingerprint or password, and times out immediately.
      • qznc 31 minutes ago
        Colleague recently told me his agent tried to sudo. It failed but it saw that docker with root is available and just used that.
    • augusto-moura 58 minutes ago
      Mainly networking and namespaces, same reasons why we run services on docker instead of old multi-user setups
    • altcognito 1 hour ago
      In a corporate environment they may not have that option
      • SoftTalker 1 hour ago
        But they can run arbitrary VMs?
    • TheRealPomax 47 minutes ago
      If you think those two contexts are equally secure, your machine is incredibly insecure and I hope you run daily incremental full system backups.

      You have far too much data in unsecured locations, and you have far too little understanding of what an agent would do, to go "I trust whatever this user account will be doing on my machine".

    • petesergeant 1 hour ago
      am I doing that once per project, or?
  • lucamartinetti 1 hour ago
    I also wrote one where you can choose to use a container or a vm, with a MITM proxy on top https://github.com/agentcage/agentcage
  • NortySpock 1 hour ago
    Have you seen https://news.ycombinator.com/item?id=48892015 and yoloai? https://github.com/kstenerud/yoloai

    Seems like both projects are following very similar approaches.

    • guywithahat 1 hour ago
      Funny that they're both written independently and in Go. Sometimes there's just a best way to do things I guess
  • rock_artist 58 minutes ago
    > Requires macOS 14+ on Apple silicon. (Linux is supported via firecracker and currently experimental...)

    So should be noted it's mostly macOS out of the box with some Linux support if I understand right.

    • celrenheit 6 minutes ago
      mostly macOS out of the box. I've tested it on linux and it worked but it's not my daily driver. I'll make the platform scope clearer in the README
  • statt 54 minutes ago
    When firing up agent VMs for my personal projects, I tried to completely separate SSH keys as well. Surprisingly annoying to have multiple keys for, say GitHub, and restrict the agent to use a deploy key.

    I saw this project enables ssh-agent forwarding, so my question; is this a non-issue to begin with? Or just not your focus currently.

  • skybrian 1 hour ago
    If the agent is running on your machine, it will suspend when you put your laptop asleep. I prefer using a remote Linux VM to let the coding agent keep working.

    I’m quite happy with exe.dev for this. My laptop is asleep upstairs but I have an agent coding away in a browser tab on the tablet I’m using. I could also check on it from my phone.

    But it might also be nice if a setup similar to exe.dev were available for self-hosting. I have a Mac Mini that I don’t really use much.

    • nzoschke 27 minutes ago
      Another +1 to cloud sandboxes (and exe.dev) vs laptop sandbox.

      Runs 24/7 completely air gapped from your laptop.

      You also want a service proxy so the sandbox can access GitHub or Stripe without keys accessible to the agent. I haven't seen many of the laptop sandbox tools do this, where exe.dev does it out of the box with their "integrations".

      I usually drop my own binary agent coding toolkit inside the sandbox so I have things like a code browsing and review right there in every sandbox too.

      https://github.com/housecat-inc/scratch

      Like you I also have a Mac Mini I've thought about making into my own 24/7 dev box, but building this vs buying 50 VMs from exe.dev for $20/mo doesn't add up for me.

    • mswphd 1 hour ago
      https://paseo.sh/ supports self-hosting, though I've only used it a mild amount tbh.
      • skybrian 37 minutes ago
        It looks like they have the right idea but their workspaces are based on GitHub worktrees rather than separate VM’s.
    • matheusmoreira 1 hour ago
      > it will suspend when you put your laptop asleep

      It is possible to simply not do that. Laptops work just fine as servers. They even have a builtin monitor and UPS.

      • skybrian 43 minutes ago
        Yeah but I ‘d rather run it remotely and not run agents on my laptop at all.
  • dietdrb 1 hour ago
    We are building the cloud version of this: hosted, isolated VMs. MacOS and Linux supported. Cloud means you can run many VMs in parallel.

    https://bitrise.io/platform/remote-dev-environments

  • windex 19 minutes ago
    I use a raspberry pi for all my experiments
  • felooboolooomba 1 hour ago
    I use mkosi: https://github.com/systemd/mkosi

    Which is just a front for systemd-nspawn. It's annoying you have to edit the config.nspawn to mount a directory if you start it with the "shell" command, instead of booting. But apart from that, it's brilliant.

  • ttul 55 minutes ago
    And, make coding harnesses run 20x faster at many multi-processing and file operations by running the coding harness itself inside of Linux instead of MacOS…
    • mlnj 54 minutes ago
      Is file IO a bottleneck now?
  • daitangio 1 hour ago
    I am developing a super light similar thing here

    https://github.com/daitangio/take-ai-control

    It is docker + vscode friendly. I tested it with major systems (copilot, codex, Claude Code and pi.dev) Comments Wellcome!

  • aftbit 1 hour ago
    Does Codex run its own sandbox? I see that sometimes it runs commands without asking, which then fail for some reason, and it asks to run them again "outside of the sandbox"
  • bad_haircut72 1 hour ago
    Sprites from Fly io does this beautifully, claude comes preinstalled, its great
    • whitlock 16 minutes ago
      The core abstraction seems identical to Fly.io Sprites: give the coding agent its own real Linux machine, put a hard network/isolation boundary around it, then yeet.

      Sprites arguably has the better security boundary, since the agent isn’t sitting adjacent to my laptop and home network.

  • c0rruptbytes 1 hour ago
    everyone loving these linux sandboxes, even Apple wrote one https://github.com/apple/containerization/tree/main/examples...
  • oleg_kabanov 26 minutes ago
    What does Clawk do that Docker without exposed ports doesn't?
  • rvz 1 hour ago
    This is like the 30th AI sandbox project on Show HN. Why this one over the rest?
    • SoftTalker 1 hour ago
      Also the readme (and thus likely the whole thing) is clearly LLM-generated.
      • mixedbit 38 minutes ago
        and probably this is why there are so many such projects. With agents, a sandbox is an important tool to have, with LLMs it is easy to vibe-code a new project, so we have dozens of sandbox projects. With so much noise, it is hard for these projects to reach a critical mass, so we may continue to see more and more sandboxes. Most will get a couple of stars on GitHub, but will be used mainly by the author. Unfortunately, we may see the same pattern with any other problem domain, dozens of projects, many vibe coded, many of questionable quality, nothing reaching maturity due to too much noise.
      • celrenheit 41 minutes ago
        [dead]
  • thiodrio 51 minutes ago
    Is it similar to devcontainers?
  • anoop_kumar 1 hour ago
    why not just use something like smol VMs? So many VM's and I am confused on which one to use for what. I think we now need a VM orchestrator!
  • prairieroadent 1 hour ago
    why not just a docker container
    • matheusmoreira 1 hour ago
      Because that means you are sharing kernel with the sandboxed agent. Virtualization presents an infinitely smaller attack surface.
      • hwc 1 hour ago
        If there is any attack surface within a properly-configured container, that's a kernel bug, right?
        • matheusmoreira 1 hour ago
          Probably. If I remember correctly, containers on Linux are implemented using the kernel's namespaces. The same ones which became famous for the vulnerabilities they surfaced in previously unexercised code.
      • croes 51 minutes ago
        Then use SmolVM
        • matheusmoreira 42 minutes ago
          I'm already using virtdev, my own solution built on top of QEMU.
  • htrp 1 hour ago
    how does this compare with the aws lambda microvms?
  • TheRealPomax 50 minutes ago
    > On your own machine that leaves two bad options. You approve every command (and babysit a prompt every few seconds), or you run --dangerously-skip-permissions and hope nothing important is one rm -rf or one leaked token away.

    Literally everyone has the option to use a VM - it's built into Windows, UTM on MacOS, Docker on Linux. Yes, "a tool that automatically builds a VM" is useful, but we've had a third option (four, if you count "actually I disagree with the idea that it's only useful if it's fully agentic") from day one.

  • ebeirne 1 hour ago
    love what youve done here. i will be using this in the future.
    • celrenheit 1 hour ago
      Thank you, that means a lot !
  • 0xbadcafebee 1 hour ago
    How many of these are there now, a hundred? We get it, you can run an agent in a VM/container/sandbox. What about configuration management & rollbacks? What about the policy engine? What about dynamic credential management? What about the lethal trifecta? A sandbox is the easiest part and doesn't address the others.
    • matheusmoreira 0 minutes ago
      Huh. My virtdev project implements nearly all of that... Except credential injection from the host, which is on my TODO list.
  • bitwize 1 hour ago
    Errbody gangsta until the agent figures out it's in a container and finds an exploit that lets it break out of container jail...
    • matheusmoreira 1 hour ago
      I developed a VM project just like this one. Asked Fable to stress test it and try to break out of containment, and to my surprise it didn't manage to. Fable didn't get downgraded to Opus either, for some reason.

      Would have thrown Mythos at it if I had access to it.

      • arcanemachiner 1 hour ago
        Isn't Fable just Mythos + guardrails? Sounds like you did throw Mythos at it.
        • matheusmoreira 1 hour ago
          I initially thought that was the case: literally the same weights but with an incredibly obnoxious "safety classifier" tacked on. Now I'm uncertain because people on HN have told me it's a different model altogether with further fine tuning for safety or something.
      • nullc 1 hour ago
        You should conduct the same test with knowingly faulty containment, otherwise the theory that the model is hobbled should probably outrank that it couldn't escape.
        • matheusmoreira 51 minutes ago
          That's a good idea. I don't have Fable at the moment, when it comes back I'll ochestrate those tests.

          Opus did manage to iron out a lot of fail open bugs during development though, and Fable's like a hundred times more relentless than Opus. I'm not saying it's a definitive result or that my VM firewall thingy is unhackable... I'm just saying it put a smile on my face.

  • mlnj 55 minutes ago
    Is there anyone who is NOT building another AI sandbox solution?
  • petesergeant 1 hour ago
    Yet another one, only I wrote this one, so I prefer it: https://github.com/pjlsergeant/byre

    You might prefer byre's simplicity, transparency, and ease of reasoning about: one local container, explicit access grants, readable generated Docker, and a workflow that stays close to normal development rather than introducing a larger sandbox platform. It's also very very easy to eject from if you want to stop using it.

  • elombn 1 hour ago
    [flagged]
  • sumar7 1 hour ago
    [flagged]
  • escape_42 53 minutes ago
    [dead]
  • mrbn100ful 1 hour ago
    YAY, let's build the same thing over and over :/

    Did you know that before reinventing the wheel, you can ask LLM your requirements, and they can find the most suited already existing tool ?

    • matheusmoreira 22 minutes ago
      Yeah, but I explicitly chose to make my own anyway because I wanted to fully own the code and none of the existing solutions worked the way I wanted.

      There is absolutely nothing wrong with reinventing the whell. It's entirely possible to do it and end up discovering you've made a better wheel.

    • 28304283409234 14 minutes ago
      `vagrant snapshot` exists.
    • croes 49 minutes ago
      Isn’t the point of LLM to recreate the wheel?

      Tl libraries, no frameworks. Let AI recreate it from scratch … every time