FRESH

Hacker News

Home

Show HN: Smooth CLI – Token-efficient browser for AI agents

58 points by antves

by tekacs

2 subcomments

This looks really interesting!
I _would_ be curious to try it, but...
My first question was whether I could use this for sensitive tasks, given that it's not running on our machines. And after poking around for a while, I didn't find a single mention of security anywhere (as far as I could tell!)
The only thing that I did find was zero data retention, which is mentioned as being 'on request' and only on the Enterprise plan.
I totally understand that you guys need to train and advance your model, but with suggested features like scraping behind login walls, it's a little hard to take seriously with neither of those two things anywhere on the site, so anything you could do to lift up those concerns would be amazing.
Again, you seem to have done some really cool stuff, so I'd love for it to be possible to use!
Update: The homepage says this in a feature box, which is... almost worst than saying nothing, because it doesn't mean anything? -> "Enterprise-grade security; End-to-end encryption, enterprise-grade standards, and zero-trust access controls keep your data protected in transit and at rest."

by oofbaroomf

2 subcomments

Is this essentially a cloud-managed specialized subagent with an LLM-friendly API?
Seems like an interesting new category.

by dandaka

0 subcomment

I'm working on building a personal assistant concept. One test I've been running is asking different AI assistants to use a browser to check available appointment slots for my hairstylist. None of them has managed to do it successfully yet, but I'm going to keep trying.
https://n694923.alteg.io/company/656492/personal/menu?o=

by jwr

1 subcomments

I was actually very interested until I realized that this doesn't run on my computer…
I get the sandboxing, etc, but a Docker container would achieve the same goals.

by sandgraham

0 subcomment

Curious how this compares to https://sentienceapi.com/. My understanding is that Sentience uses deterministic "semantic snapshots" to try and give agents a more reliable browser interface.

by tobyhinloopen

1 subcomments

Way too expensive, I'll wait for a free/open source browser optimized to be used by agents.

by ilaksh

1 subcomments

This is a good idea. Do you use something like browser-use or Fara-7b behind the scenes? Or maybe you don't want to give up your secrets (which is fine if that's the case).

by itissid

0 subcomment

How does your approach differ from BrowserOS, not in the product sense(their product is ane enterprise browser based off chrome). but in how they designed the interface between the browser and the models?

by KevinChasse

0 subcomment

Interesting approach. Exposing high-level goals rather than UI actions definitely reduces token overhead, but reproducible comparisons with open-source setups would strengthen the claim. Also, remote browsers introduce a new attack surface—sandboxing helps, but I’d like to see clear isolation guarantees against malicious pages or rogue scripts.

by lasgawe

1 subcomments

I'm a bit curious. Why did you link the docs instead of the site in this post?

by quotemstr

0 subcomment

Interesting idea as an open source tool I could hack locally, but no way in hell am I adding yet another bill and using a web browser of all things as SaaS. I'll make my own web-specialized subagent.

by waynenilsen

1 subcomments

Frontend QA is the final frontier, good luck, you are over the target.
The amount of manual QA I am currently subjected to is simultaneously infuriating and hilarious. The foundation models are up to the task but we need new abstractions and layers to correctly fix it. This will all go the way of the dodo in 12 months but it'll be useful in the meantime.
agent-browser helped a lot over playwright but doesn't completely close the gap.

by franze

1 subcomments

Congrats for shipping.
How does it compare to Agent Browser by Vercel?

by waynenilsen

2 subcomments

i can see a new token efficient mirror web possibly emerging using content type headers on the request side
forms, PRG, semantic HTML and no js needed

by behnamoh

1 subcomments

Ironically, the landing page and docs pages of Smooth aren't all that token-efficient!

by desireco42

1 subcomments

Look this is cool idea, but subscribing to anything these days is a hard sell, we are all tired of subscription plans. You probably would be more succesful if you could find this to package in a way that is not subscription.