awayto · HackerTrans

il y a 7 moisdiscuss

Both have 24/7 twitch channels

https://www.twitch.tv/rifftrax

https://www.twitch.tv/mst3k

il y a 8 moisdiscuss

> Run pwd to see the directory you’re working in. You’ll only be able to edit files in this directory.

If you're using the agent to produce any kind of code that has access to manipulate the filesystem, may as well have it understand its own abilities as having the entirety of CRUD, not just updates. I could easily see the agent talking itself into working around "only be able to edit" with its other knowledge that it can just write a script to do whatever it wants. This also reinforces to devs that they basically shouldn't trust the agent when it comes to the filesystem.

As for pwd for existing projects, I start each session running tree local to the part of the project filesystem I want to have worked on.

awayto

il y a 8 moisdiscuss

Could you point to some resources which talk about how docker isn't considered a safe sandbox given the network and file system restrictions I mentioned?

I understand the sharing of kernel, while I might not be aware of all of the implications. I.e. if you have some local access or other sophisticated knowledge of the network/box docker is running on, then sure you could do some damage.

But I think the chances of a whitelisted llm endpoint returning some nefarious code which could compromise the system is actually zero. We're not talking about untrusted code from the internet. These models are pretty constrained.

awayto

il y a 8 moisdiscuss

Google has what I would call a generous free tier, even including Gemini 2.5 Pro (https://ai.google.dev/gemini-api/docs/rate-limits). Just get an API key from AiStudio. Also very easy to just make a switch in your agent so that if you hit up against a rate limit for one model, re-request the query with the next model. With Pro/Flash/Flash-Lite and their previews, you've got 2500+ free requests per day.

awayto

il y a 8 moisdiscuss

Yeah I agree. Ultimately I would suggest not having any kind of function call which returns an arbitrary command.

Instead, think of it as if you were enabling capabilities for AppArmor, by making a function call definition for just 1 command. Then over time suss out what commands you need your agent do to and nothing more.

awayto

il y a 8 moisdiscuss

You can build your agent into a docker image then easily limit both networking and file system scope.

    docker run -it --rm \
      -e SOME_API_KEY="$(SOME_API_KEY)" \
      -v "$(shell pwd):/app" \ <-- restrict file system to whatever folder
      --dns=127.0.0.1 \ <-- restrict network calls to localhost
      $(shell dig +short llm.provider.com 2>/dev/null | awk '{printf " --add-host=llm-provider.com:%s", $$0}') \ <-- allow outside networking to whatever api your agent calls
      my-agent-image

Probably could be a bit cleaner, but it worked for me.

awayto

il y a 8 moisdiscuss

Also if you're doing function calls you can just have the command as one response param, and arguments array as another response param. Then just black/white list commands you either don't want to run or which should require a human to say ok.

awayto

il y a 3 ansdiscuss

    Location: Oregon
    Remote: Yes
    Willing to relocate: No
    Technologies: recently I'm using Javascript/Typescript, Node, SQL, Docker, Shell -- but I've been developing for 25 years and used a lot of stuff.
    Resume: On request. https://github.com/jcmccormick 
    Email: joe \at\ keybittech.com

My name is Joe. I am a software engineer. I build my own software, Awayto, which you can currently demo at https://awayto.store to see my output. The demo features a communications platform, similar to what you would expect if you were running an online writing center. There is RBAC, scheduling & online appointments, voice/video calling, shared whiteboard/document marking, and more. However, the system is built in a generic fashion where features can be added or removed on a whim, leading to a pivotable system for whatever business you're creating.

I'm interested in working with businesses with the above feature set interests, educational groups, community services, and so on. The software is fully open source and runs bare metal or (currently) on hetzner. As a freelancer I am looking for work on projects of similar scope, as well as implement Awayto for your own purposes, should you need.

I've been developing for 25 years and am of course open to other projects as well. And I look forward to hearing what ideas you have, and your needs.

Please contact me at joe \at\ keybittech.com for any inquiries, and check my profile for more info and links to my work.

comments