134 points by antoniomika 3 days ago | 117 comments | View on ycombinator
saagarjha 3 days ago |
ryanczak 3 days ago |
__MatrixMan__ 3 days ago |
I guess I could do the same thing with VM's, but for reasons I can't fully explain it's just easier to stay vigilant with separate hardware.
yevpats about 7 hours ago |
knocte 3 days ago |
Good approach. I myself found issues with a VM approach such as latency (if the VM is in the cloud), efficiency (if the VM is local) and laziness to automate the VM creation & setup. So instead I just hand the agent an unprivileged user account in my system and stop worrying. I built a little tool to setup this: https://github.com/tarsgate/skynot
h4kunamata 3 days ago |
Tokens in general is "never exposed":
keepassxc-cli show --no-password -k $KEEPASS_KEY $KEEPASS_DB "GROUP_GOES_HERE/SECRET_NAME_HERE" --attributes Password
When dealing with agents, they access the tokens the same way, it is the best balance I found between keeping things safe without adding too much complexity making my life harder.
It can save tho, have the code to also save access_token to the vault the same way, so no log/execution will ever show the secrets.
Tokens the agents use, are read-only with the minimalist access.
Never ever ever grant an agent access to any token with destructive access, unless you accepted the risks.
LeBit 3 days ago |
Smolvm does that.
I make use of that feature which is super neat.
One problem with that though is when you need to use such a credential on a tailscale node. It bypasses the proxy of those tools.
blop 3 days ago |
If you run for example `fnox exec gh`, fnox will ask you to touch the yubikey to decrypt the age key and decode the github token env var for the gh cli (fnox can also cache the decoded value for a bit, like gpg agent). You can tell the AI agent to use `fnox exec gh` to access gh without seeing the gh token. Or create a shell alias so it's transparent. It also has an MCP server and proxy server to filter API calls to remote APIs (I haven't tried that yet)
it supports quite a few providers (https://fnox.jdx.dev/providers/overview.html): age, fido2, AWS, Azure, Google, kubernetes secrets, vault/openbao, 1password, bitwarden, keepassxc, etc.
A lot of them are read/write so you can also use fnox to sync secrets across different systems.
tmpz22 3 days ago |
Thank you for posting and sharing this OP, its one small step in the direction of professional development with LLMs.
zmmmmm 3 days ago |
rao-v 3 days ago |
It's actually a pretty interesting problem and I'm hoping to share what I've thrown together this week.
Basically you want to be able to limit access to a bunch of resources to stuff you authorize (from your phone or inside Herdr etc.) - everything from access to specific websites, search APIs, keys to your Home Assistant config ("Claude make an automation to turn off the lights when blah" is magic), to github tokens and wifi passwords etc. so that the agents themselves never see the actual key (ok not quite possible with wifi passwords but everything else is possible!).
At the same time your setup has to be simple enough that fairly dumb agents can figure it out without burning a billion tokens on every call.
LittleCloud 3 days ago |
Beyond SSH key forwarding as described in TFA, I’ve also made (with agents of course) a filtering HTTP(S) proxy that injects Authorization credentials, so LLMs operating REST APIs (including commands like Github’s gh) don’t see my private keys either even if they try to hack my environment (which is sandboxes). Anybody else doing that?
munishjindal931 1 day ago |
kstenerud 3 days ago |
By default:
- The agent has no access to your filesystem.
- The agent has no access to your env.
- The agent has no access to your credentials.
- The agent has no access to your workdir. You use git semantics to pull its changes back to your workdir, with the ability to view and extract it as a diff.
- You can restrict where it's allowed to visit over the network.
Requests can be brokered (still working on making this more ergonomic), and you can make explicit carve-outs from the above rules.
qudat 3 days ago |
cyanydeez 3 days ago |
Still figuring out how it all works, but ephemeral one time pads are going to make a good comeback, I assume.
ball_of_lint 3 days ago |
holtwick 3 days ago |
undefined 3 days ago |
recroad 3 days ago |
theyCallMeSwift 3 days ago |
bob1029 3 days ago |
Perhaps my prompts are extremely low risk and I've been incredibly lucky. What are the statistics on 100k+ powershell invokes without a single attempt to delete system32? Perhaps I'm overdue for a black swan event? In any case, I'm looking forward to it because it would finally make all this discussion seem rational rather than completely unhinged.
yoyohello13 3 days ago |
bismanawaz 3 days ago |
scotty79 3 days ago |
sooraj_sj 3 days ago |
charlie2agent 3 days ago |
ofirmakmal 3 days ago |
maxkaya 3 days ago |
gma692002 3 days ago |
victorylaunches 3 days ago |
tokenemargin 3 days ago |
runtarmac 3 days ago |
xiaoliangliang 3 days ago |
tushar_kr 3 days ago |
abapforagents 3 days ago |
sick_of_slop 3 days ago |
uain 3 days ago |
tulo 3 days ago |
Traubenfuchs 3 days ago |
All it needs me for is to pass two factor gates. It can pull aws-cli secrets and whatever else it needs from our SSO portal after I signed in in the morning.
The desire to maintain an illusion of control over AI which is already much faster and smarter than you is hilarious to me.
Just tell it to confirm before doing writing/destroying operations (send money, send email, create kubernetes cluster, destroy sns/sqs, etc.) and you are good.