dbmikus

Are there any good guides on how to write prompt files tailored to different agents?

Would also be interested in examples of a CLAUDE.md file that works well in Claude, but works poorly with Codex.

Every website needs to add the "friend or foe" system[0] so that I can mark bots to avoid their content and mark good posters so I can filter just to theirs.

[0]: https://hackersmacker.org/

From the article, it looks like they integrated with Docker because someone at Docker reached out about collaborating on the integration.

Regarding security, I think you need three things:

  1. You need the agent to run inside a sandbox.
  2. You need a safe perimeter or proxy that can apply deterministic filtering rules on what makes it into the AI agent's sandbox and the HTTP requests and responses that agent sends out from the sandbox.
  3. The bot should have its own email accounts, or maybe be configured to only send/read from certain email addresses

I'm working on a product that makes it as easy to spin up remote agent sandboxes as it is to git push and git pull. Then when we get that working well we're putting a proxy around each sandbox to let users control filtering rules.

I personally see a future where there are many different types of *Claws, coding agents, etc. and I think they need a new "operating system", so to speak.

Self-plug at the end: https://github.com/gofixpoint/amika. The OSS part of my startup, focused on sandbox coding agents right now :)

PS: I enjoyed the entropytown.com blog! bookmarking it

Cool, thanks for explaining!

Appreciate the deets!

It sounds like a big part of your use case is to safely give an agent control of your computer? Like, for things besides codegen?

We're probably not going to directly support that type of use case, since we're focused on code-gen agents and migrating their work between localhost and the cloud.

We are going to add dynamic filesystem mounting, for after sandbox creation. Haven't figured out the exact implementation yet. Might be a FUSE layer we build ourselves. Mutagen is pretty interesting as well here.

I like that it's all bash.

How does this compare with Codex's and Claude's built-in sandboxing?

I've been working on an OSS project, Amika[1], to quickly spin up local or remote sandboxes for coding workloads. We support copy-on-write semantics locally (well, "copy-and-then-write" for now... we just copy directories to a temp file-tree).

It's tailored to play nicely with Git: spin up sandboxes form CLI, expose TCP/UDP ports of apps to check your work, and if running hosted sandboxes, share the sandbox URLs with teammates. I basically want running sandboxed agents to be as easy as `git clone ...`.

Docs are early and edges are rough. This week I'm starting to dogfood all my dev using Amika. Feedback is super appreciated!

FYI: we are also a startup, but local sandbox mgmt will stay OSS.

[1]: https://github.com/gofixpoint/amika

> Or mainly to save the end user the hassle to set it up correctly

It's this.

Don't have a Dropbox moment ;) [1]

[1]: https://news.ycombinator.com/item?id=9224

I think the README could use a few real use-case examples. I understand it in an abstract sense, but not sure I understand the benefit vs plain-text communication, besides saving on token spend.

I'm going to be pedantic and note that iOS and Android both have the capability security model for their apps.

And totally agree that instead of reinventing the wheel here, we should just lift from how operating systems work, for two reasons:

1. there's a bunch of work and proven systems there already

2. it uses tools that exist in training data, instead of net new tools

The people that are good but dangerous drivers will drive well and safely during tests, so you won't catch them.

You can market products that people need. A big part of this is explaining and educating someone about what your product does, another part is just getting the word out there. Every website homepage is more-or-less a marketing page.

If no one is marketing a product, then nobody knows about it.

What kinds of data are you working on? Coding? Something else?

I've been curious how much these AI models look for more niche coding language expertise, and what other knowledge frontiers they're focusing on (like law, medical, finance, etc.)

Check out Org-Roam: https://www.orgroam.com/

It has some Obsidian-like features inside Org Mode.

So, if you're looking for an easier-to-use UI, it's not it, but if you're looking for Obsidian-like linking and backlinking, it has that.

This is cool. A bunch of interesting things here:

  1. Agent can create its own tools and save them to memory
  2. You create a SQL (and web app?) workbench per agent run
  3. Grok fell off a cliff in the last month. Was this consistent over multiple runs?
  4. Agents have a difficult time backtracking. Would unwinding system state and agent context make backtracking better? (Harder to implement this, though)
  5. Since each new month only uses final state from previous month, agent has no way to understand why error occurred in previous month

Cool experiment! Was it difficult building the observable SQL workbench? And how many humans-in-the-loop did you have?

I'll hop in there!

This is awesome! How much of your team's time goes into working on the physical hardware, versus RL simulation environments, versus managing all the training data from the real robot and the simulations?

I'm super interested in learning more about the training process of world and robotics model and the data challenges there.

Many bots use residential IP proxy networks, so they come from the same IPs that humans use

What vendors or platforms are frontier labs moving to?

VC industry likes to mention that there are bad actors without actually mentioning the bad actors. It's a faux pas to actually name and shame. The industry happens to be small enough and super relationship-driven so, for better or worse, observing the faux pas matters.

When should I use this versus Browserbase or Browserless or Hyperbrowser?

Obviously since this is open source, then I can self host it. What other reasons?

Just curious!

This is BS that Trump is making up so he can say the USA is at war and deport people without due process.

https://www.reuters.com/world/us/us-judge-extends-block-trum...

If the goal is safety, rewriting ancient battle-tested software is not the way to do it.

The Lindy Effect for software: the longer software appears to be bug-free, the more likely it is bug-free.

The alternative is that the starter pricing tiers are more expensive. Pick your poison.

I feel like every enterprise ad or marketing site is like this. I have no idea what it does, but it says it'll make numbers go up, so execs buy. It must work to a degree, because it's so common.

I never want to claim certainties, but it seems pretty close to certain that symbolic AI loses to statistical AI.

I think there is room for statistical AI to operate symbolic systems so we can better control outputs. Actually, that's kind of what is going on when we ask AI to write code.

100% agree. When we explicitly segment and compose AI components, we are removing the ability for them to learn their own pathways between the components. We've been proven time and time again the bitter lesson[1]: that throwing a ton of data and compute at a model yields better results than what we could come up with.

That said, we can still isolate and modify parts of a network, and combine models trained for different tasks. But you need to break things down into components after the fact, instead of beforehand, in order to get the benefits of learning via scale of data + compute.

[1]: http://www.incompleteideas.net/IncIdeas/BitterLesson.html

Hey, I'm not OP, but I'm working on what seems to be the exact problem you mentioned. We (https://fixpoint.co/) search and monitor web data about companies. We are indexing patents and academic papers right now, plus we can scrape and monitor just about any website (some social media sites not supported).

We have users with very similar use cases to yours. Want to email me? dylan@fixpoint.co. I'm one of the founders :)

Just did a pricing check, and you give twice as much concurrency/browsing/data-transfer as Browserbase. Nice!

Do you have any restrictions on sites we can scrape? I am likely going to do some public LinkedIn scraping (logged out, legally compliant).

Congrats on the launch! Definitely a needed product. BTW, your docs link is broken, but working docs link is here: https://www.humanlayer.dev/docs/introduction