Show HN: Browser AI agent platform designed for reliability
55 points
15 hours ago
| 9 comments
| github.com
| HN
We’re very excited to share something we’ve been building. Notte https://www.notte.cc/ is a full-stack browser agent platform built to reliably automate a wide range of workflows.

Browser agents aren’t new, but what is still hard is covering real-world flows reliably. The inspiration for Notte was to make a full-featured platform that bridges the agent reliability gap. We’ve packaged everything via a singe API for ease of use:

- Site Interactions - Observe website states, scrape data and execute actions

- Structured Output - Get data in your exact format with Pydantic models

- Stealth browser sessions - built-in CAPTCHA solving, proxies, and anti-detection

- Hybrid workflows - Combine scripting and AI agents to reduce costs and improve reliability

- Secrets vaults - Credential management to store emails, passwords, MFA tokens, SSO, etc.

- Digital personas - Digital identities with unique emails, phones for account creation workflows

With these tools, Notte allows you to automate difficult tasks like account creation, form filling, work on authenticated dashboards. Close compatibility with Playwright allows you to cut LLM costs and improve execution speed by mixing web automation primitives and include agents only for specific parts that require reasoning and adaptability.

Here’s a short YouTube demo: https://www.youtube.com/watch?v=b1CzmfpdzaQ

If any of this sounds interesting, you can run your first agent following our quickstart on GitHub https://github.com/nottelabs/notte. Or play around with our free plan through our Notte Console: https://console.notte.cc/

We’d love to hear if there’s anything else required before you’d try or trust it on your own workflows :)

xena
5 hours ago
[-]
Do you support robots.txt so administrators can opt-out of your service?
reply
ogandreakiro
2 minutes ago
[-]
[delayed]
reply
jaflo
13 hours ago
[-]
The pricing page mentions how many credits you get but not what a credit does or gets you. Could you elaborate on that?
reply
anonzzzies
1 hour ago
[-]
But with AI it never tells you that. I pay $xxxx for the Claude API, half of the time it breaks it's own code and just does things explicitly told it NOT to do; I think I want to be refunded for the time wasting and frustration, but that's not happening. So you pay credits and you get 'stuff' that 'might work'; that's the reality. I'm gonna say this is basically not even allowed for consumers in the EU, but I don't know that for sure.
reply
ogandreakiro
10 hours ago
[-]
Oh thanks for pointing this out. Pricing is explained on our landing https://notte.cc.

It’s 500 free credits. Pro plan $79/mo for 10K credits/mo. You can do additional top up 1K credits for $10 when in Pro plan.

If you need more it’s a custom enterprise deal - we work a lot with volume discounts.

reply
doesnt_know
7 hours ago
[-]
It's simple! For $50 a month we give you a thousand smeckles which you can use to crank the floba. A very good deal.
reply
GoatInGrey
9 hours ago
[-]
To reiterate, "what a credit does or gets you" is not explained by your pricing page.

I'll operate on the assumption that it takes roughly 500 credits to scrape a single page for now.

reply
Bluestein
10 hours ago
[-]
Just want to congratulate all involved for the obvious amount of care that has gone into thinking this through ...

I really hope it succeeds through the roof.-

reply
ogandreakiro
10 hours ago
[-]
Thanks - that’s nice to hear! :)
reply
Bluestein
9 hours ago
[-]
I've got to say - and this in a way makes sense: The model seems to be very "URL-oriented" ...

... as in (at least in the demo, the search one) if you ask it to visit a site by name (ie. "Go go Altavista and ...," ...

... it will complain "no URL" has been provided.-

It probably has to do with the "special instructions" given to the AI for this one program/demo ...

... still.-

reply
defied
14 hours ago
[-]
Looks good, excited to try this together with the CDP feature. If it works well we can add it to our (TestingBot) integration/documentation.
reply
ogandreakiro
13 hours ago
[-]
Hey! Sure happy to create a shared integration with TestinBot - shoot me an email so we can discuss andrea@notte.cc :)
reply
monoid73
15 hours ago
[-]
for the hybrid workflows, curious how do you decide which parts need AI reasoning vs can be hardcoded? is it adaptive or manual config?
reply
ogandreakiro
13 hours ago
[-]
right now it's hardcode - you write a web automation script where at some parts you have a very specialized agent to take the very few steps in your workflow that require reasoning/adaptability. Future: we're trying to make this process automatic.
reply
revskill
2 hours ago
[-]
Credit based pricing is broken. No way u produce shiy then user get out of credit to get nothing done.
reply
meowtimemania
2 hours ago
[-]
There needs to be a way to purchase "X" more credits then it works IMO
reply
Jommi
14 hours ago
[-]
just tried to use it to extract data from hyatt.com

completely failed

another hype but actually doesn't work browser agent.

reply
ogandreakiro
14 hours ago
[-]
hey - thanks for trying out. not sure what you're trying to extract from hyatt.com but just naively tested the most basic agent (https://github.com/nottelabs/notte?tab=readme-ov-file#run-in...) with simple prompt task="scrape data from hyatt.com landing page" and it worked perfectly:

answer=Successfully navigated to hyatt.com and closed the cookie consent. The page content includes navigation links (Explore, Offers, Meetings & Events, Loyalty Program, Language, Sign In or Join), a search form with inputs for City/Address/Landmark/Airport/Hotel, Check-in and Check-out dates, and options for rooms/guests and special rates. Below the search form, there are sections for 'Earn bonus points at new hotels', 'Limited-Time Offer: Earn 2 Free Nights + 25,000 Bonus Points', 'More Places to Discover' with specific hotel examples (Park Hyatt Auckland, Hyatt Regency London - The Churchill, 7Pines Resort Ibiza, Hyatt Centric Murano Venice), 'Join World of Hyatt' with benefits, and various promotional sections like 'Incredible offers, incredible places', 'Plan a summer of more', 'Outdoor adventure awaits', and 'Wellbeing your way'. The page also lists different Hyatt brands categorized into Luxury, Lifestyle, Inclusive, Classics, and Essentials. At the bottom, there are links for Reservations, Customer Service, Corporate Sites, Explore More, and Connect with Hyatt (social media links), along with legal information (Privacy Policy, Terms & Conditions, Cookie Center, Security & Safety, Modern Slavery and Human Trafficking, Do Not Sell or Share My Personal Information)

reply
alstonite
10 hours ago
[-]
How well does it handle captchas?
reply
ogandreakiro
10 hours ago
[-]
There’s no magic here unfortunately. We’re working hard to make anti-detection as good as possible with proxies to avoid captchas in first place. When they get there - we can solve ~60% of providers right now (incl reCAPTCHA, Cloudflare, and main ones) and some others are still work in progress.
reply
zekdk
10 hours ago
[-]
does your scraping endpoint also provide stealth mode?
reply
ogandreakiro
10 hours ago
[-]
Yes! We just added stealth support for our scraping endpoint. The docs should be updated soon but you can set proxies now
reply