10 ms·

Show HN: Uruky (EU-based Kagi alternative) now has Image Search and URL Rewrites

You can get a 2h free trial by solving a proof-of-work captcha when topping up your account for the first time. If you'd like to learn more, an independent interview was posted a couple of weeks ago [1], and the FAQ [2] has a lot of information as well. For the source code sharing, we've talked with lawyers and are inclined to no longer require the NDA/NCC for privacy reasons shared with us before (signing requires identification), but instead use a source-available permissive license that doesn't allow competition, like PolyForm Shield [3] (we do still have about 6 months before finalising a decision, here). This does come with a lot more risks for us (it's harder to track down if someone publishes the code or uses it against the license), but given we've already passed 100 monthly active accounts, we're feeling more confident it's an acceptable risk. The plan is to give logged in accounts (who are 12 months old or more) a way to download a ZIP of the current code base that's in the server. Obviously there's no easy way to prove that's the case, but we're open to ideas/suggestions if someone here has them. [1]: https://theprivacydad.com/interview-with-the-engineer-of-uruky-a-private-search-engine/ https://theprivacydad.com/interview-with-the-engineer-of-uru... [2]: https://uruky.com/faq https://uruky.com/faq [3]: https://polyformproject.org/licenses/shield/1.0.0 https://polyformproject.org/licenses/shield/1.0.0

kreyenborgi 5mo ago
Do you pronounce it like uruk hai?
- BrunoBernardino 5mo ago
  It's in the FAQ, we pronounce it like "Euro-key" in English.
Munksgaard 5mo ago
Interesting! All the license stuff aside, there's definitely a desire for more EU-first services like this.
danielrmay 5mo ago
Does Uruky provide an API, or allow API usage such a way that I can leverage it as part of an agent workflow, or otherwise, in place of something like DuckDuckGo?
- amarcheschi 5mo ago
  There should be according to their faq https://uruky.com/faq https://uruky.com/faq
- BrunoBernardino 5mo ago
  Yes, we do offer an API and are transparent about the rate limits (no search limits) in the FAQ.
ainiriand 5mo ago
It is a bit hard to evaluate the potential when you need to top up and do a captcha just for evaluation purposes. The barrier of entry is quite high.
- BrunoBernardino 5mo ago
  To be clear, you don't need to top up (only click the link or try a search) _and_ do the captcha. You can _either_ top up for a month, _or_ do the proof-of-work captcha (should just be clicking a checkbox and giving it a few seconds). The reason for the higher barrier of entry is bots/abuse. Up until a few weeks ago we didn't even have the captcha option. Without any kind of identification, it's impossible to prevent automated/bot signups, and they can abuse the system to oblivion. I understand that's not ideal, though I hope it still makes sense!
  - iLoveOncall 5mo ago
    I'm not making a freaking account before I can even run a single search.
    - Munksgaard 5mo ago
      As far as I can tell, Kagi also requires you to create an account before you can try it out. You can of course disagree with that practice, but it is not like Uruky is requiring something far out of the ordinary.
    - inquirerGeneral 5mo ago
      [dead]
    - BrunoBernardino 5mo ago
      To create an account you only need to click a link (so the account number can be randomly generated for you), not provide any information. I understand if that's not good enough for you, though.
  - aniviacat 5mo ago
aniviacat 5mo ago
> EU search providers (Marginalia, Mojeek, EUSP, etc.). Does this mean this is just a meta search engine without its own index? If so, the comparison to Kagi seems misleading. The question would turn from "why not Kagi" to "why not SearxNG".
- freehorse 5mo ago
  Kagi is also a metasearch engine, as it combines several third-party indexes.
- SyneRyder 5mo ago
  Kagi is also a meta search engine. The only Kagi-owned index is Teclis, which is very small and really only indexes RSS feeds of small sites. The Kagi API supplements its Teclis results with additional results from Marginalia. It would be nice to have more search indexes available though, especially via APIs. I wish Kagi was a search index, back when I was a Kagi subscriber that's what I hoped my funds were going towards building.
- BrunoBernardino 5mo ago
  Good question! We also have Uruky Site Search, our own index, which is quite tiny at the moment, but we hope to grow, eventually. Kagi's own index is Teclis, which is one of their smaller providers as well.
  - aniviacat 5mo ago
    Sounds good, thanks for the clarification.
  - SyneRyder 5mo ago
    If you are building out your own index, might you consider offering it as an API via pre-paid credits (ala Mojeek, Kagi Teclis?) I'm only paying Mojeek about $10 - $20 per year in API for my personal metasearch, so I guess this is a terrible market to enter ;) But I'd genuinely be interested, especially if the money is going towards building an index.
danielspace23 5mo ago
I think that, more than EU metasearch engines, we need EU search indexes. EUSP is already something, but they seem to be working rather slowly, compared to how quickly Brave built their own index. It's also trivial to run a perfectly working metasearch engine with the same sources as Uruky, it's called Searxng. In any case, good luck on this project. I personally don't think it's for me. Maybe a better user interface would change the equation, but as of now I'll stick to Qwant.
- BrunoBernardino 5mo ago
  Yes! Uruky Site Search is our own EU-based index and we'd love to expand it, but that takes an enormous amount of resources (which we don't yet have). Besides EUSP we also allow you to choose and use Mojeek and Marginalia, two other big EU-based indexes, and by using Uruky with them, you're supporting them financially, directly.
- carlosjobim 5mo ago
  Suggested name: Memory hole.
evilmonkey19 5mo ago
I really like the idea and that it's eu-made a love it. A fee things I see with kagi which are useful and improvements: - Hire a UI/UX person NOW! My parents and gf like using google and kagi because are easy to use. - add the widgets like the football or the show the local store with the phone number asap. My gf is thinking about moving away from kagi because of this. - the quick ai response is extremely useful. - Indexing websites is super important. People doesnt know where to put the content in a website or how to make accessible. Many times i use google due to this fact. - Make a family subscription. - make it funny, easy to use and welcoming. The branding is SUPER important. Good luck and I really wish you to succeed! Im paying for an account ;)
- BrunoBernardino 5mo ago
  Thank you for your suggestions and support! - Ouch. We get a lot of love for our UI/UX, which I guess just goes to show taste is personal. If you have any particular points of friction, we'd love to hear about them. - Only Serper allows for this, so we're very hesitant. If other providers allowed it, we could definitely consider it. - We won't have Generative AI, sorry. - Could you clarify? - Check out the interview I gave The Privacy Dad, I explain the problems for something like that. - Roger!
  - wolvoleo 5mo ago
    Big plus for not doing unrequested AI stuff <3
  - lowdude 5mo ago
    Seeing the point on UI/UX got me curious as well and from a brief look I don't quite see where it is coming from, but maybe they will expand on it. Regarding their point about relatives loving Google/Kagi for usability: If you don't already do that, I believe it can be pretty eye-opening to just place some "tech-illiterate" person in front of your product and check if/where they struggle. There were lots of situations where I saw people struggling with steps I stopped thinking about, just because I was more familiar with the software. And while I have nothing to do with UI/UX myself, I am pretty sure that kind of paper cut is what tends to block any kind of broad adoption.
my_throwaway23 5mo ago
As much as I'd prefer a smaller, non-american operator (for most anything really), I'm extremely hesitant to pay directly for search. For this service, the "just an ID as account" looks nice and private on the surface, but once you look at payment methods, it's 100% personally identifiable. If it's so privacy-focused - where's the payment option for transferring Monero? As for the code - don't get me started. Source available? NDAs? Smells "Private VPN" funny to me.
- BrunoBernardino 5mo ago
  We're working towards accepting cash by mail in a month or two, and regarding cryptocurrencies, we've been trying to reach out to proxysto.re unsuccessfully for that. If you know of a similar option, we'd love to hear about it!
  - embedding-shape 5mo ago
    If the goal truly is privacy, and that's why you're adding cryptocurrency options, then adding a middleman like "proxysto.re" or any other, then you're not giving users the privacy you say you want to give. You really need to handle that on your end, if that's something you want to offer for the purpose of privacy. Take a look at how Mullvad implemented it, which I guess is where you got the "cash in a letter" method from, they also handle cryptocurrencies by themselves for exactly this reason, there is no way for you to actually be able to fulfill the privacy promise otherwise.
    - BrunoBernardino 5mo ago
      Yes, and I’ve talked with Mullvad about it before. The main problem for us is that we cannot exchange XMR for fiat legally, in Portugal, and we have no costs that can be paid with XMR, at the moment. ProxyStore should work well as a proxy because they receive XMR anonymously and pay us in EUR for an account number. We don’t know who paid for what and don’t care (we also don’t know that after 14 days, Mollie doesn’t know any account number).
Cider9986 5mo ago
I will pay for search once Kagi or Brave accepts Monero and offers a reasonable price. Can I pay for Uruky with Monero? If you're worried about regulations consider reaching out to proxysto.re
- BrunoBernardino 5mo ago
  Thanks! I have reached out over a week ago to them via email and Signal, to crickets. Let me know if you know why or how to reach them, or alternatives.
  - Cider9986 5mo ago
    Hmm, they responded to me on Signal quickly at one point so unsure.
    - BrunoBernardino 5mo ago
      Thanks. They did respond after a few days, we're in touch with them now and will try to run a trial selling Uruky vouchers there!
cromka 5mo ago
I hate to say this, but those recent EU alternative counteroffers for popular products bare some really unfortunate names....
- BrunoBernardino 5mo ago
  From our FAQ: > The name has no special meaning but we read it like "Euro-key" in English. Names are hard, and we're aware it can remind people of Uruk and Uruk-hai. That's OK. I completely understand your point, though. It's not "Google"!
  - _ache_ 5mo ago
    "Google", for french speaking people, looks like "Gogol" which means "stupid" or "retarded".
alex7o 5mo ago
I live in the EU and I use kagi, but I think you fail to understand why kagi is good and useful. In the end of the day I need a search engine to find stuff and kagi is better than google for the things I and my AI agents are searching for. If you don't get that no amount of better privacy is worth my time as a person as shit searches mean agents don't works so well a.k.a more expensive or my room becomes toastier for the agents that I run locally. (I can use it for boring personal searches, but I do 2/3 of these a day, so I am not paying 5€ for that)
- scrollaway 5mo ago
  Agreed. The noise in tech circles often gets founders to conflate ten different things into a product that no longer makes sense. “Eu made alternative to Kagi”? Cool, we need European search engines, sign me up. “Privacy is such a priority we’re looking to accept cash by mail”? Okay, you’re never gonna build a serious competitor, never mind.
  - wolvoleo 5mo ago
    Yeah Mullvad that accept cash in the post are not a serious VPN provider at all, right.
    - scrollaway 5mo ago
      You seem to not understand that a search engine and a VPN don't have the same audience, and certainly not the same needs for focus. It's ok, we all have our flaws.
      - wolvoleo 5mo ago
        They are very comparable from a privacy standpoint IMO. A search engine and a VPN both get quite a lot of insight in your interests and browsing habits because a lot of browsing sessions start with a search.
- NoboruWataya 5mo ago
  Have you tried Uruky and found it to be lacking? Totally agree with you, but I don't see many people here saying they've had a bad experience with this. (I haven't used it)
carlosjobim 5mo ago
I think this is excellent so that Kagi can focus on making the best search engine instead of trying to please the angriest and most difficult customers.
wolvoleo 5mo ago
Interesting. Being EU-based is a huge plus over Kagi in my opinion. I also like that they don't ask any personal data, even email address. I like services that don't want any personal details. Like with Mullvad, where they just give you a random number and that's your account ID <3 Unfortunately Mullvad enshittified in other ways so I had to move to ProtonVPN. But services that act like that are great IMO. Unfortunately a lot of services apply "Know your Customer" BS even though they are not in the financial sector. However I wonder where they get their search data from. But it's worth investigating.
- BrunoBernardino 5mo ago
  It should be clear in the FAQ. All Europe-based. Our current search providers: - Mojeek - EUSP - Marginalia - Linkup - Serper (the only non-EU, being UK and proxying Google) - Uruky Site Search (our own index) - Pixabay (images) I'd be curious to hear about your Mullvad experience (feel free to email me).
  - wolvoleo 5mo ago
    Well Mullvad stopped providing port forwarding, which is necessary for good torrent support (especially if you use public torrents and you need ones that aren't seeded a lot). They also stopped supporting openvpn which I need. Wireguard only now. It's not enshittification in terms of ads etc but it is reduction of possibilities because they already make enough money on the people using the main features. But anyway I rely on both things so I moved over.
    - KomoD 5mo ago
      What's the reason you need OpenVPN?
      - wolvoleo 5mo ago
        I use a hardware router with really nice encryption offloading that does not support wireguard but does support openvpn natively. This router is in front of my 'arrrrr' subnet at home, where all my leeching servers live. This way sidechannel attacks don't work because they have no alternative path to the internet. It's a nice setup and I have no wish to change it or to buy other hardware.
bigblind 5mo ago
It doesn't do great on recent events it seems. The amusement park Walibi Belgium recently announced a company called RMC is doing a makeover of their wooden roller coaster, so I did a search for "walibi belgium rmc", and it found one very out-of-date article about earlier rumours, and a bunch of less relevant stuff.
- BrunoBernardino 5mo ago
  Thanks! I'm currently in the middle of building a date filter I expect to launch later today. Still, was this using Mojeek or EUSP? Have you tried Linkup or Serper for those kinds of searches? You don't have to change your default search providers, you can just choose one from the bottom of the results list.
  - lolc 5mo ago
    Wait do I have to choose what search provider I use for every search? They don't get merged by Uruky?
    - BrunoBernardino 5mo ago
      Only if they don’t provide enough results. We initially queried them all and merged, but that became too costly and inefficient (too many duplicates most times). It should be explained in the FAQ.
axegon_ 5mo ago
I'd be more than willing to subscribe and support the project BUT, I need to address the elephant in the room: The reason why I'm against Kagi is the fact that they use Yandex(be it only for images according to their own words) and I'm sure as hell refusing to give a single cent to them. So I guess my question is: sources?
- psychoslave 5mo ago
  What's wrong with Yandex? Is it quality of service being too low? Does the company have tracking records of shady practices, worst than what GAFAM are doing that make customer flee away?
  - wolvoleo 5mo ago
    It's Russian. Indirectly you are supporting Putin when you use it. Those sanctions aren't for no reason.
    - axegon_ 5mo ago
      ^ Correct!
    - chadgpt3 5mo ago
      Am I indirectly supporting Trump when I use DuckDuckGo?
      - woadwarrior01 5mo ago
        Yes. You're also supporting Bing ads. With a thin veneer of privacy washing goodness, of course.
      - dodslaser 5mo ago
        Yes, and a lot of EU citizens/companies are actively in the process of migrating away from US based software/products to EU alternatives for, among others, this exact reason.
      - 5mo ago
asciimoo 5mo ago
As a search/metasearch developer, I can heavily recommend Uruky. It is by far the best third party search alternative in terms of how they approach privacy and transparency. Keep up the good work Bruno! <3
- BrunoBernardino 5mo ago
  Thank you so much, Adam! Hister [1] is also an amazing option for people looking to have their own "local" index! And it's FOSS! [1] : https://hister.org https://hister.org
janandonly 5mo ago
I read in your FAQ that you don't support bitcoin payments yet. Might I suggest you have a look at https://www.lightspark.com/capabilities/embedded-finance https://www.lightspark.com/capabilities/embedded-finance ? They make it easy to be compliant with local laws and integrate payment features.
- BrunoBernardino 5mo ago
  Thanks! We're currently trying to get in touch with proxysto.re for that. Light Spark seems based in the US, so we can't consider them, I'm afraid.
  - chadgpt3 5mo ago
    Why do you need to involve a third party to implement bitcoin payments?
    - BrunoBernardino 5mo ago
      Not for bitcoin, but for monero, which is more requested than bitcoin.
      - chadgpt3 5mo ago
        Why do you need a third party?
        BrunoBernardino 5mo ago
        It's in the FAQ. Legal limitations.
dizhn 5mo ago
Do you have Privacy Pass or some other type of privacy-preserving credentials functinoality?
- BrunoBernardino 5mo ago
  Can you expand on the benefits of that when there is no personal information (not even an email or payment ID after 14 days) associated with an account?
  - dizhn 5mo ago
    I was just assuming this will be a subscription service like Kagi.
    - BrunoBernardino 5mo ago
      Gotcha.
Alex_toani 5mo ago
It's a cool search engine. I agreee that need improve in UI/UX. If UI/UX is great. I will consider to use it.
- BrunoBernardino 5mo ago
  Thanks, if you have any specific points of friction, please let us know (an email is perfectly fine)!
sarjann 5mo ago
I think it could help to maybe allow 10 searches free (without logging in) just for someone to see what the UI is like or show an example of the results page. It adds a bit of friction for someone to have to make an account and pay just to test the product. Kagi lets you test it offline (go incognito and try) " Kagi Search is funded by members, not advertisers: built to find what you need, not sell your attention. Try 50 free searches, and if you love it, sign up for 100 more before choosing a plan. Searches used "
- BrunoBernardino 5mo ago
  Thanks for the suggestion! That would mean we'd have to count searches, which we don't want to do. With the 2h free trial via proof-of-work captcha it should allow you for much more than that! I know the FAQ for that is outdated! I'll update it later today with a new date filter.
sthix 5mo ago
Regarding UI/UX. One thing that immediately catches my as an Design Engineer (being over 7 years Product Designer before that) is that the site looks bland. Nothing that sticks. Also one UX caveat: you are using your accent-color for primary buttons like the sign-up as well as a display variant for text for the "PRIVATE SEARCH YOU CONTROL". The logo is not a strong one, lacks personality, the font choice is also a little "weird". Just my two cents. But I am glad, that someone is creating an alternative in the EU. Hit me up, if you want to get more design opinions.
mrngld 5mo ago
Honestly didn't notice Kagi was US-domiciled. Their meetups I think are typically in Europe and their co-working and office space is in Belgrade. Anyway, Kagi's excellent, their search results for me are significantly better than Google (and customizable), they've leveraged AI in a way thats optional and, to me, class-leading in its ability to help with search. "But we're EU based" is a product strategy that might land you some local government contracts and a few customers whose key motivation is negative emotion towards others but it's never going to be the path to great success. Spotify didn't conquer the globe because they framed themselves as an anti-US anti-iMusic or anti-Pandora or whatever alternative. They conquered because their product was solid. Nobody cared where it's HQ was.
- NoboruWataya 5mo ago
  It seems clear to me that a company that pushes being EU-based and minimising US dependencies as a selling point doesn't really want to "conquer the globe". As a consumer it is even less important to me. Indeed, if your search provider has conquered the globe, then it's probably time to find a new one because the next logical step is turn the screws on users to maximise profits.
- BrunoBernardino 5mo ago
  Great point! In our comparison page with Kagi it should become clearer that’s not our only product differentiator. If you’re happy with Kagi, that’s great, it’s a very good product! We don’t want world domination. :)
Kudos 5mo ago
Country filters for Luxembourg and Monaco, no country filter for Ireland?
- BrunoBernardino 5mo ago
  Oops! Sorry about that. I’ll add it later today.
KomoD 5mo ago
> EU servers, EU storage, EU payment processing, EU search providers (Marginalia, Mojeek, EUSP, etc.). > All servers and data are physically in the EU. All search providers are based in the EU. Payment processing is done in the EU. Mojeek and Serper aren't EU so that's just false. And I'm not sure all the providers only use EU servers so I don't like the claim for that reason too. > Try in: Google // DuckDuckGo // Ecosia I would remove this, I thought it was for changing what provider I was using, but no, it just sent me to Google. Then at the bottom of the page is where I found this, which I would prefer to have at the top. > Try with: Mojeek // EUSP // Linkup // Serper
- BrunoBernardino 5mo ago
  Dang. You’re right that Mojeek is UK, so Europe but not EU (at the moment, at least). I’ll need to add a remark for that. We already do it for Serper. Regarding the servers, our servers and data are in the EU. We can’t guarantee that for our providers, but they don’t access your data. As for Try in/Try with, good point, we can add a preference for swapping them.
localhoster 5mo ago
Yavos - Yet Another Vibe-cOded Service :)
Hackbraten 5mo ago
Have you considered adding Google’s SERP to your indexes (via some third-party provider)? Google’s index is by far the largest, and my impression is that a search engine is hardly useful unless it includes Google’s results.
- BrunoBernardino 5mo ago
  Thanks! Serper is UK-based and does that. We provide them as an option, but not by default.
s_dev 5mo ago
Does Uruky pay Yandex like Kagi does? It's relevant to those of us boycotting Russian products and influence due to the Ukraine war.
- BrunoBernardino 5mo ago
  The search providers we use are listed in our FAQ and all are Europe-based. Yandex is not included.
dethos 5mo ago
Interesting. I will definitely add it to my list of services to experiment with. I like the account/payment system, where you top up a random account number for a period. Instead of having to go with the whole process of creating and verifying an account with your data and then managing yet another recurring subscription. Congrats on the release, wish you the best.
- BrunoBernardino 5mo ago
  Thank you! Please reach out if you have any suggestions or run into any trouble!
blfr 5mo ago
Can I (and how to) use it as a search engine for Open WebUI?
- BrunoBernardino 5mo ago
  I don't know what that is and it seems like some AI dashboard. Uruky has an API so as long as you keep within the rate limits, you should be fine!
AndroTux 5mo ago
Okay, so in order to do a test search, I have to: 1) type in a query and hit enter because there's no search button. 2) click signup, even though I want to evaluate it before creating an account. 3) apparently now I'm signed up without having to enter any details - what's the point? Just create a new session as soon as I initiate the search. 4) so now I need to return to the homepage to trigger another search. 5) search again, enter again. Now I'm greeted by a captcha. 6) after solving the captcha, I now have to enter my search query a third time because it wasn't saved 7) search results! Guys.
- BrunoBernardino 5mo ago
  The goal was to always show you the account number before you went for anything, but your post made me realize you don't need that if you're just trying it out, so now you should only: 1) type in a query and hit enter 2) solve the captcha 3) search results!
  - AndroTux 5mo ago
    Much better. Now just add a freaking search button. That's literally your primary function, and you're lacking the button to do the one thing. I understand as a power user you don't press on buttons, but from a UX perspective, if you don't trigger the search immediately upon typing, you need a button to start the search. Your current UI suggests: Type in a query, and then click on "images" because that's the next thing that you can click. (Also apparently my IP is now banned and now every page just reads "Forbidden," so I couldn't even buy your product if I tried. You really should invest a lot in UX if you want this to be a product that's used by anyone other than hardcore privacy nerds.)
Havoc 5mo ago
Nice - I like this. Especially that I can also use it via API too and that expectations are limits are set out clearly (which seem very generous). On limits - consider changing the short limit on this: > searches to 1 per second, 30 per minute, and 1800 per hour. to 5 per 5 seconds or 10/10. That still works out to one per second but users are less likely to accidentally hit it with two requests that have similar timing. Say one from me and one from API usage.
- BrunoBernardino 5mo ago
  Good point. We have never had any reports of customers legitimately hitting these, but we can probably loosen them up a bit.
  - Havoc 5mo ago
    Yeah can't picture the average user getting anywhere near the 2nd and 3rd limit. They seem very high to me, but the short one I can definitely picture hitting easily under API scenarios. e.g. Suppose you have 3 subagents getting triggered that all want to do a search. >can probably loosen them up a bit. Definitely wouldn't move the top 2. That would just be hard to walk back later
  - Havoc 5mo ago
    I see this has already been updated in the FAQ . Thanks!
    - BrunoBernardino 5mo ago
      Thank you for the suggestion!
yegle 5mo ago
Given that: - Google has been around for 20+ years, so the concept of search engine and the technologies behind it should have been well known. - Computing power and Internet speed has increased significantly and in many homes (at least outside of US) 1Gbps is norm. - Everyone is talking about Google deteriorating over the years and prefer the old Google. The old system from the 2000s should be dirt cheap to run with modern home hardware. - People's need for search engine is highly specific, you presumably would be interested in searching a small subset of the whole Internet. My question is: why haven't a local run search engine be a thing at least in the tech circles? It should be able to bootstrap with e.g. an hourly updated "top 100 websites in 50 categories" index file, and adapt to my daily queries to automatically update the index in the background, and iteratively improves the quality of the results. The rise of the local LLM users proves this model works.
- BrunoBernardino 5mo ago
  You should check out hister.org!
  - yegle 5mo ago
    Oh it checked multiple boxes I have in mind, and the website UI is really great. I'll definitely give it a try.
- randomtoast 5mo ago
  kagi exists for quite some time
  - yegle 5mo ago
    I was talking about a self-hosted search engine.
ent 5mo ago
Please don't try to machine translate to my local language. The translations are terrible or outright incorrect.
- BrunoBernardino 5mo ago
  Thanks for reporting this. We’ve used a professional translator for German (though we’ve added many new phrases that haven’t gone through that process since), and we speak Spanish and French (besides English and Portuguese). We use DeepL for the rest but it’s not perfect. What’s your language, and do you know of a professional to help us? If you’d rather reach out to us via email, it also works.
dtj1123 5mo ago
One of the things I like about Mullvad is that I can buy a voucher on Amazon, and use that to pay for the service with a high degree of anonymity. Is/will something like that be possible with Uruky?
- BrunoBernardino 5mo ago
  I’m not aware of this, can you clarify? Technically anyone can top up an account on Uruky and give it away to someone else.
  - dtj1123 5mo ago
    Like so, hopefully the link works for you: https://amzn.eu/d/00yFWx41 https://amzn.eu/d/00yFWx41 The card arrives with a little lottery-ticket style scratchpad with I believe a 16 digit activation code underneath.
    - BrunoBernardino 5mo ago
      Thanks, I see. You're still giving Amazon and Mullvad your payment information, so unless you're purchasing it for someone else, I'm failing to see the advantage? Again, that's something you can already do with Uruky as well.
      - dtj1123 5mo ago
        The advantage is that it makes it impossible for Mullvad or Amazon to associate my online payment with the activation code on the card, and by extension my Mullvad account.
        BrunoBernardino 5mo ago
        Gotcha. We're now building vouchers for ProxyStore to sell them.
- john_strinlai 5mo ago
Yizahi 5mo ago
Hi. A have a user report, after visiting the site it automatically switched to a language based on my system locale, but not on my UI language or keyboard preference. So it is possible to visit the site from some guest PC for example where locale would be something unfamiliar. And the language selector is down at the very bottom (I guarantee that my mom for example would never find it) it is plain text and it is in the currently set language. So for example if I will get a page set to Chinese or Korean or Hindi, I wouldn't even understand where language selector is, it would be a set of glyphs unrecognizable to me. Or vice versa, let's say an Indian person would get a cyrillic text. I suggest moving language selector to the header of the page and adding some icons to it - flags or English encodings (EN/DE/CN/RU/IN): https://en.wikipedia.org/wiki/List_of_ISO_639_language_codes https://en.wikipedia.org/wiki/List_of_ISO_639_language_codes
- BrunoBernardino 5mo ago
  Thanks for reporting this. Do you happen to know what your browser sends in the `Accept-Language` header?
  - Yizahi 5mo ago
    I think this is what is sent from my side: "name": "Accept-Language", "value": "en-US,uk-UA;q=0.9,uk;q=0.8,en;q=0.7" "name": "User-Agent", "value": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:151.0) Gecko/20100101 Firefox/151.0"
    - BrunoBernardino 5mo ago
      Thanks, I've pushed a fix for this!
      - Yizahi 5mo ago
        Works on my side now, thank you. ;)
        5mo ago
maxloh 5mo ago
Could you consider Delayed Open Source Publication (DOSP) [1]? With that, you make the source code open-source after a specific period of time (such as 5 years). This way, your customers still get the source code and can use it freely, but you don't have to worry about competition. By the time anyone could use it to compete with you, they would be using an ancient version of your software. The BUSL (Business Source License) by MariaDB is a battle-tested license designed for exactly this use case. [1]: https://opensource.org/delayed-open-source-publication https://opensource.org/delayed-open-source-publication
- BrunoBernardino 5mo ago
  That is an option, for sure, but it feels less honest to our customers to give them the code "so much later". With the PolyForm Shield, they can just use the _current_ version however they want for themselves/family, as long as they're not competing with us commercially. Or am I missing something else, here?
  - maxloh 5mo ago
    Not really. That means you give your customers the code to use under your current restrictions, and after a specific period, they can use it under the open-source license of your choice as well.
    - BrunoBernardino 5mo ago
      I see. We'll look into considering doing both, then. PolyForm Shield for the current code, with BUSL to transition into AGPLv3 after 2 or 3 years. Do you think that would work well?
      - maxloh 5mo ago
        Yeah. That would be great!
        BrunoBernardino 5mo ago
npodbielski 5mo ago
> You can use the f=json search query parameter to get JSON-formatted search results, as per the FAQ entry above. Nice so EU-based company with API available for my agent to use. Right now I am using lightpanda but this my be simpler. I will check it out. Thanks.
zuzululu 5mo ago
I just tried to do a search and it immediately asks me to create an account??? I'm just not seeing the connect here I understand you are trying to fight spam/abuse but seems like a tough uphill battle especially when VPNs solve the privacy issue entirely. I'm also not sure why this obsession with EU is so prevalent and misguided. EU is privacy regulated, not a guarantee of privacy. GDPR is better than what many places have, but hosting in UE doesn’t magically mean the data is safe, it doesnt mean 0 knowledge encryption, no insider access, no breaches, no subpoenas, no natsec carveouts, or uniformly enforced across borders.
- BrunoBernardino 5mo ago
  "Create an account" is simply generating a random account number, but I _just_ pushed a fix to make that "auto-happen" for you if you try to simply search — you'll be presented with an option to solve a captcha or pay for access, then get redirected back to the search after that. The "EU obsession" for us is mostly about trying to buy and stimulate "the local economy" as much as possible, but YMMV and you're right it doesn't mean any of those things.
frangonf 5mo ago
Happy customer here. Found you last month while looking for searxng and related alternatives for API search with clankers in a personal setting.
- BrunoBernardino 5mo ago
  Thank you for supporting us!
prism56 5mo ago
I've just cancelled Kagi. While it was great. I'm trying to stop supporting US companies unless they provide good value
Havoc 5mo ago
uhm, I'm suddenly completed locked out? Can't even get to the homepage in private browse anymore so must be IP level. Just says "Forbidden". Must have tripped some sort of safeguard while trying to figure out how to use the API via curl
- Havoc 5mo ago
  OK this is a bit ridiculous. The FAQ does not specify how to use the API beyond listing a json parameter. If you send a curl command with that parameter the first search just returns "Logged In", and run another search and then you're temporarily IP banned site wide. Having hair trigger sensitive security around API while not telling the user what the request should look like to not trigger that is just silly.
  - Havoc 5mo ago
    OK bit more trial and error later I worked out a curl command that doesn't trip this. https://pastebin.com/BNRBfYFf https://pastebin.com/BNRBfYFf Still not entirely sure what about the original request trigger it. Think it was something around the Sec-Fetch-* headers that copying a request from browser via copy cURL included
    - BrunoBernardino 5mo ago
      Sorry! You can use the incognito token (from the incognito URL) in order to not have to send the cookie over. If you were trying too aggressively to search, you could've been rate-limited for a few minutes.
      - Havoc 5mo ago
        Good to know about the incognito toke - helpful. Yes, I saw the 3 auth try mentioned in the faq after this, that must have been what I was hitting with the experimenting. Anyway all good now
Marciplan 5mo ago
As a European-first kinda person, I think we're good with Kagi?
- BrunoBernardino 5mo ago
  Kagi is a great product if you're happy with them! Uruky is for people who want ad-free, private, and personalized search, and no AI, while stimulating the European economy. That's not everyone and that's fine!
mietek 5mo ago
My number one request: let me block specific sites from my search results.
- 1kurac 5mo ago
  FYI, on reasonable search engines like altpower.app, you can exclude sites with `-site:domain.tld` keyword. See also: https://altpower.app/#Advanced%20Search%20Operator%20Syntax https://altpower.app/#Advanced%20Search%20Operator%20Syntax https://www.exploit-db.com/google-hacking-database https://www.exploit-db.com/google-hacking-database
- BrunoBernardino 5mo ago
  `-site:domain.tld` works for "non-permanent blocking", but in your settings you can add domains or TLDs to block from the search results.
senko 5mo ago
Kagi's only office/hub is in Belgrade, which may not be EU, but it's (literally) close enough. Employees are remote. Freediver (founder) is US based and Kagi is an US entity, so must comply with any warrants there. But I guess they could set up a Serbian subsidiary?
Melatonic 5mo ago
Have you considered pivoting or prioritizing different test markets or niches? For example someone brought up making your UI and whatnot easier to use for someones "grandparents". What if you included a "grandpa" mode (you dont need to call it that specifically) that simplies everything and makes it super straight forward to use. I can see technically inclined people buying a subscription for their aging families to help them avoid clicking on useless (or dangerous) crap. You could also do something similar to Kagi SmallWeb but specifically for EU based stuff. Currently the UI is not bad but for some reason throws off my BS filter. I think it might be the spacing and (maybe) lines between the results - it sort of makes it resemble the paid text ads inserted into Google Search?
manytimesaway 5mo ago
Nice project! It lacks two things that are very useful : website blocking, and !hashbangs like DDG.
- BrunoBernardino 5mo ago
  I'm not sure if this is trolling, but we offer both. Even custom bangs, and you can also use Kagi's Bangs: https://github.com/kagisearch/bangs https://github.com/kagisearch/bangs
  - manytimesaway 5mo ago
    This was not trolling at all, I didn't see these features on the website. Amazing, thanks for the quick response!
iqandjoke 5mo ago
Tried searching its name uruky on Uruky, some NSFW results shows up which I did not expected...
nutifafa 5mo ago
login gated with a paywall of 5 dollars, could not use the product.
theamk 5mo ago
It would be super cool if you chose 20-100 queries (both common and niche) and shown the sample search results for those. Those would be completely cached/pre-rendered, so the cost would be very low and there will be no potential for abuse. And yet the visitors could at least get some idea about the quality of your search engine.
dukisamardzic 5mo ago
Solid EU-first positioning — the "trust local infrastructure" angle is increasingly important here. Curious about the privacy model: Kagi famously refuses to log queries. What's your approach to search history retention?
- BrunoBernardino 5mo ago
  No logging or tracking as well.
expedited123 5mo ago
Consider adding favicons near the websites in search results. Way easier to distinguish them at a glance. --- Also: The EUSP index is surprisingly good, by the way. Bingish level from my experience...
plextoria 5mo ago
This intrigued me and I topped up an account. Is there no password? Anyone could "steal" my account number and kick me out?
- BrunoBernardino 5mo ago
  There is no password, just like Mullvad. You should "protect" your account number like you would a password, but also, you can just discard it and change to another one in a month, for example.
  - plextoria 5mo ago
    Thanks for your answer! Makes me curious how do you defend against people simply bruteforcing or guessing the account numbers.
    - BrunoBernardino 5mo ago
      Rate limiting when it's invalid (and high entropy to generate new ones)! It's in the FAQ :)

kreyenborgi 5mo ago

Do you pronounce it like uruk hai?

BrunoBernardino 5mo ago
It's in the FAQ, we pronounce it like "Euro-key" in English.

Munksgaard 5mo ago

Interesting! All the license stuff aside, there's definitely a desire for more EU-first services like this.

danielrmay 5mo ago

Does Uruky provide an API, or allow API usage such a way that I can leverage it as part of an agent workflow, or otherwise, in place of something like DuckDuckGo?

amarcheschi 5mo ago
There should be according to their faq https://uruky.com/faq https://uruky.com/faq
BrunoBernardino 5mo ago
Yes, we do offer an API and are transparent about the rate limits (no search limits) in the FAQ.

ainiriand 5mo ago

It is a bit hard to evaluate the potential when you need to top up and do a captcha just for evaluation purposes. The barrier of entry is quite high.

BrunoBernardino 5mo ago
To be clear, you don't need to top up (only click the link or try a search) _and_ do the captcha. You can _either_ top up for a month, _or_ do the proof-of-work captcha (should just be clicking a checkbox and giving it a few seconds). The reason for the higher barrier of entry is bots/abuse. Up until a few weeks ago we didn't even have the captcha option. Without any kind of identification, it's impossible to prevent automated/bot signups, and they can abuse the system to oblivion. I understand that's not ideal, though I hope it still makes sense!
- iLoveOncall 5mo ago
  I'm not making a freaking account before I can even run a single search.
  - Munksgaard 5mo ago
    As far as I can tell, Kagi also requires you to create an account before you can try it out. You can of course disagree with that practice, but it is not like Uruky is requiring something far out of the ordinary.
  - inquirerGeneral 5mo ago
    [dead]
  - BrunoBernardino 5mo ago
    To create an account you only need to click a link (so the account number can be randomly generated for you), not provide any information. I understand if that's not good enough for you, though.
- aniviacat 5mo ago

aniviacat 5mo ago

> EU search providers (Marginalia, Mojeek, EUSP, etc.). Does this mean this is just a meta search engine without its own index? If so, the comparison to Kagi seems misleading. The question would turn from "why not Kagi" to "why not SearxNG".

freehorse 5mo ago
Kagi is also a metasearch engine, as it combines several third-party indexes.
SyneRyder 5mo ago
Kagi is also a meta search engine. The only Kagi-owned index is Teclis, which is very small and really only indexes RSS feeds of small sites. The Kagi API supplements its Teclis results with additional results from Marginalia. It would be nice to have more search indexes available though, especially via APIs. I wish Kagi was a search index, back when I was a Kagi subscriber that's what I hoped my funds were going towards building.
BrunoBernardino 5mo ago
Good question! We also have Uruky Site Search, our own index, which is quite tiny at the moment, but we hope to grow, eventually. Kagi's own index is Teclis, which is one of their smaller providers as well.
- aniviacat 5mo ago
  Sounds good, thanks for the clarification.
- SyneRyder 5mo ago
  If you are building out your own index, might you consider offering it as an API via pre-paid credits (ala Mojeek, Kagi Teclis?) I'm only paying Mojeek about $10 - $20 per year in API for my personal metasearch, so I guess this is a terrible market to enter ;) But I'd genuinely be interested, especially if the money is going towards building an index.

danielspace23 5mo ago

I think that, more than EU metasearch engines, we need EU search indexes. EUSP is already something, but they seem to be working rather slowly, compared to how quickly Brave built their own index. It's also trivial to run a perfectly working metasearch engine with the same sources as Uruky, it's called Searxng. In any case, good luck on this project. I personally don't think it's for me. Maybe a better user interface would change the equation, but as of now I'll stick to Qwant.

BrunoBernardino 5mo ago
Yes! Uruky Site Search is our own EU-based index and we'd love to expand it, but that takes an enormous amount of resources (which we don't yet have). Besides EUSP we also allow you to choose and use Mojeek and Marginalia, two other big EU-based indexes, and by using Uruky with them, you're supporting them financially, directly.
carlosjobim 5mo ago
Suggested name: Memory hole.

evilmonkey19 5mo ago

I really like the idea and that it's eu-made a love it. A fee things I see with kagi which are useful and improvements: - Hire a UI/UX person NOW! My parents and gf like using google and kagi because are easy to use. - add the widgets like the football or the show the local store with the phone number asap. My gf is thinking about moving away from kagi because of this. - the quick ai response is extremely useful. - Indexing websites is super important. People doesnt know where to put the content in a website or how to make accessible. Many times i use google due to this fact. - Make a family subscription. - make it funny, easy to use and welcoming. The branding is SUPER important. Good luck and I really wish you to succeed! Im paying for an account ;)

BrunoBernardino 5mo ago
Thank you for your suggestions and support! - Ouch. We get a lot of love for our UI/UX, which I guess just goes to show taste is personal. If you have any particular points of friction, we'd love to hear about them. - Only Serper allows for this, so we're very hesitant. If other providers allowed it, we could definitely consider it. - We won't have Generative AI, sorry. - Could you clarify? - Check out the interview I gave The Privacy Dad, I explain the problems for something like that. - Roger!
- wolvoleo 5mo ago
  Big plus for not doing unrequested AI stuff <3
- lowdude 5mo ago
  Seeing the point on UI/UX got me curious as well and from a brief look I don't quite see where it is coming from, but maybe they will expand on it. Regarding their point about relatives loving Google/Kagi for usability: If you don't already do that, I believe it can be pretty eye-opening to just place some "tech-illiterate" person in front of your product and check if/where they struggle. There were lots of situations where I saw people struggling with steps I stopped thinking about, just because I was more familiar with the software. And while I have nothing to do with UI/UX myself, I am pretty sure that kind of paper cut is what tends to block any kind of broad adoption.

my_throwaway23 5mo ago

As much as I'd prefer a smaller, non-american operator (for most anything really), I'm extremely hesitant to pay directly for search. For this service, the "just an ID as account" looks nice and private on the surface, but once you look at payment methods, it's 100% personally identifiable. If it's so privacy-focused - where's the payment option for transferring Monero? As for the code - don't get me started. Source available? NDAs? Smells "Private VPN" funny to me.

BrunoBernardino 5mo ago
We're working towards accepting cash by mail in a month or two, and regarding cryptocurrencies, we've been trying to reach out to proxysto.re unsuccessfully for that. If you know of a similar option, we'd love to hear about it!
- embedding-shape 5mo ago
  If the goal truly is privacy, and that's why you're adding cryptocurrency options, then adding a middleman like "proxysto.re" or any other, then you're not giving users the privacy you say you want to give. You really need to handle that on your end, if that's something you want to offer for the purpose of privacy. Take a look at how Mullvad implemented it, which I guess is where you got the "cash in a letter" method from, they also handle cryptocurrencies by themselves for exactly this reason, there is no way for you to actually be able to fulfill the privacy promise otherwise.
  - BrunoBernardino 5mo ago
    Yes, and I’ve talked with Mullvad about it before. The main problem for us is that we cannot exchange XMR for fiat legally, in Portugal, and we have no costs that can be paid with XMR, at the moment. ProxyStore should work well as a proxy because they receive XMR anonymously and pay us in EUR for an account number. We don’t know who paid for what and don’t care (we also don’t know that after 14 days, Mollie doesn’t know any account number).

Cider9986 5mo ago

I will pay for search once Kagi or Brave accepts Monero and offers a reasonable price. Can I pay for Uruky with Monero? If you're worried about regulations consider reaching out to proxysto.re

BrunoBernardino 5mo ago
Thanks! I have reached out over a week ago to them via email and Signal, to crickets. Let me know if you know why or how to reach them, or alternatives.
- Cider9986 5mo ago
  Hmm, they responded to me on Signal quickly at one point so unsure.
  - BrunoBernardino 5mo ago
    Thanks. They did respond after a few days, we're in touch with them now and will try to run a trial selling Uruky vouchers there!

cromka 5mo ago

I hate to say this, but those recent EU alternative counteroffers for popular products bare some really unfortunate names....

BrunoBernardino 5mo ago
From our FAQ: > The name has no special meaning but we read it like "Euro-key" in English. Names are hard, and we're aware it can remind people of Uruk and Uruk-hai. That's OK. I completely understand your point, though. It's not "Google"!
- _ache_ 5mo ago
  "Google", for french speaking people, looks like "Gogol" which means "stupid" or "retarded".

alex7o 5mo ago

I live in the EU and I use kagi, but I think you fail to understand why kagi is good and useful. In the end of the day I need a search engine to find stuff and kagi is better than google for the things I and my AI agents are searching for. If you don't get that no amount of better privacy is worth my time as a person as shit searches mean agents don't works so well a.k.a more expensive or my room becomes toastier for the agents that I run locally. (I can use it for boring personal searches, but I do 2/3 of these a day, so I am not paying 5€ for that)

scrollaway 5mo ago
Agreed. The noise in tech circles often gets founders to conflate ten different things into a product that no longer makes sense. “Eu made alternative to Kagi”? Cool, we need European search engines, sign me up. “Privacy is such a priority we’re looking to accept cash by mail”? Okay, you’re never gonna build a serious competitor, never mind.
- wolvoleo 5mo ago
  Yeah Mullvad that accept cash in the post are not a serious VPN provider at all, right.
  - scrollaway 5mo ago
    You seem to not understand that a search engine and a VPN don't have the same audience, and certainly not the same needs for focus. It's ok, we all have our flaws.
    - wolvoleo 5mo ago
      They are very comparable from a privacy standpoint IMO. A search engine and a VPN both get quite a lot of insight in your interests and browsing habits because a lot of browsing sessions start with a search.
NoboruWataya 5mo ago
Have you tried Uruky and found it to be lacking? Totally agree with you, but I don't see many people here saying they've had a bad experience with this. (I haven't used it)

carlosjobim 5mo ago

I think this is excellent so that Kagi can focus on making the best search engine instead of trying to please the angriest and most difficult customers.

wolvoleo 5mo ago

Interesting. Being EU-based is a huge plus over Kagi in my opinion. I also like that they don't ask any personal data, even email address. I like services that don't want any personal details. Like with Mullvad, where they just give you a random number and that's your account ID <3 Unfortunately Mullvad enshittified in other ways so I had to move to ProtonVPN. But services that act like that are great IMO. Unfortunately a lot of services apply "Know your Customer" BS even though they are not in the financial sector. However I wonder where they get their search data from. But it's worth investigating.

BrunoBernardino 5mo ago
It should be clear in the FAQ. All Europe-based. Our current search providers: - Mojeek - EUSP - Marginalia - Linkup - Serper (the only non-EU, being UK and proxying Google) - Uruky Site Search (our own index) - Pixabay (images) I'd be curious to hear about your Mullvad experience (feel free to email me).
- wolvoleo 5mo ago
  Well Mullvad stopped providing port forwarding, which is necessary for good torrent support (especially if you use public torrents and you need ones that aren't seeded a lot). They also stopped supporting openvpn which I need. Wireguard only now. It's not enshittification in terms of ads etc but it is reduction of possibilities because they already make enough money on the people using the main features. But anyway I rely on both things so I moved over.
  - KomoD 5mo ago
    What's the reason you need OpenVPN?
    - wolvoleo 5mo ago
      I use a hardware router with really nice encryption offloading that does not support wireguard but does support openvpn natively. This router is in front of my 'arrrrr' subnet at home, where all my leeching servers live. This way sidechannel attacks don't work because they have no alternative path to the internet. It's a nice setup and I have no wish to change it or to buy other hardware.

bigblind 5mo ago

It doesn't do great on recent events it seems. The amusement park Walibi Belgium recently announced a company called RMC is doing a makeover of their wooden roller coaster, so I did a search for "walibi belgium rmc", and it found one very out-of-date article about earlier rumours, and a bunch of less relevant stuff.

BrunoBernardino 5mo ago
Thanks! I'm currently in the middle of building a date filter I expect to launch later today. Still, was this using Mojeek or EUSP? Have you tried Linkup or Serper for those kinds of searches? You don't have to change your default search providers, you can just choose one from the bottom of the results list.
- lolc 5mo ago
  Wait do I have to choose what search provider I use for every search? They don't get merged by Uruky?
  - BrunoBernardino 5mo ago
    Only if they don’t provide enough results. We initially queried them all and merged, but that became too costly and inefficient (too many duplicates most times). It should be explained in the FAQ.

axegon_ 5mo ago

I'd be more than willing to subscribe and support the project BUT, I need to address the elephant in the room: The reason why I'm against Kagi is the fact that they use Yandex(be it only for images according to their own words) and I'm sure as hell refusing to give a single cent to them. So I guess my question is: sources?

psychoslave 5mo ago
What's wrong with Yandex? Is it quality of service being too low? Does the company have tracking records of shady practices, worst than what GAFAM are doing that make customer flee away?
- wolvoleo 5mo ago
  It's Russian. Indirectly you are supporting Putin when you use it. Those sanctions aren't for no reason.
  - axegon_ 5mo ago
    ^ Correct!
  - chadgpt3 5mo ago
    Am I indirectly supporting Trump when I use DuckDuckGo?
    - woadwarrior01 5mo ago
      Yes. You're also supporting Bing ads. With a thin veneer of privacy washing goodness, of course.
    - dodslaser 5mo ago
      Yes, and a lot of EU citizens/companies are actively in the process of migrating away from US based software/products to EU alternatives for, among others, this exact reason.
    - 5mo ago

asciimoo 5mo ago

As a search/metasearch developer, I can heavily recommend Uruky. It is by far the best third party search alternative in terms of how they approach privacy and transparency. Keep up the good work Bruno! <3

BrunoBernardino 5mo ago
Thank you so much, Adam! Hister [1] is also an amazing option for people looking to have their own "local" index! And it's FOSS! [1] : https://hister.org https://hister.org

janandonly 5mo ago

I read in your FAQ that you don't support bitcoin payments yet. Might I suggest you have a look at https://www.lightspark.com/capabilities/embedded-finance https://www.lightspark.com/capabilities/embedded-finance ? They make it easy to be compliant with local laws and integrate payment features.

BrunoBernardino 5mo ago
Thanks! We're currently trying to get in touch with proxysto.re for that. Light Spark seems based in the US, so we can't consider them, I'm afraid.
- chadgpt3 5mo ago
  Why do you need to involve a third party to implement bitcoin payments?
  - BrunoBernardino 5mo ago
    Not for bitcoin, but for monero, which is more requested than bitcoin.
    - chadgpt3 5mo ago
      Why do you need a third party?
      - BrunoBernardino 5mo ago
        It's in the FAQ. Legal limitations.

dizhn 5mo ago

Do you have Privacy Pass or some other type of privacy-preserving credentials functinoality?

BrunoBernardino 5mo ago
Can you expand on the benefits of that when there is no personal information (not even an email or payment ID after 14 days) associated with an account?
- dizhn 5mo ago
  I was just assuming this will be a subscription service like Kagi.
  - BrunoBernardino 5mo ago
    Gotcha.

Alex_toani 5mo ago

It's a cool search engine. I agreee that need improve in UI/UX. If UI/UX is great. I will consider to use it.

BrunoBernardino 5mo ago
Thanks, if you have any specific points of friction, please let us know (an email is perfectly fine)!

sarjann 5mo ago

I think it could help to maybe allow 10 searches free (without logging in) just for someone to see what the UI is like or show an example of the results page. It adds a bit of friction for someone to have to make an account and pay just to test the product. Kagi lets you test it offline (go incognito and try) " Kagi Search is funded by members, not advertisers: built to find what you need, not sell your attention. Try 50 free searches, and if you love it, sign up for 100 more before choosing a plan. Searches used "

BrunoBernardino 5mo ago
Thanks for the suggestion! That would mean we'd have to count searches, which we don't want to do. With the 2h free trial via proof-of-work captcha it should allow you for much more than that! I know the FAQ for that is outdated! I'll update it later today with a new date filter.

sthix 5mo ago

Regarding UI/UX. One thing that immediately catches my as an Design Engineer (being over 7 years Product Designer before that) is that the site looks bland. Nothing that sticks. Also one UX caveat: you are using your accent-color for primary buttons like the sign-up as well as a display variant for text for the "PRIVATE SEARCH YOU CONTROL". The logo is not a strong one, lacks personality, the font choice is also a little "weird". Just my two cents. But I am glad, that someone is creating an alternative in the EU. Hit me up, if you want to get more design opinions.

mrngld 5mo ago

Honestly didn't notice Kagi was US-domiciled. Their meetups I think are typically in Europe and their co-working and office space is in Belgrade. Anyway, Kagi's excellent, their search results for me are significantly better than Google (and customizable), they've leveraged AI in a way thats optional and, to me, class-leading in its ability to help with search. "But we're EU based" is a product strategy that might land you some local government contracts and a few customers whose key motivation is negative emotion towards others but it's never going to be the path to great success. Spotify didn't conquer the globe because they framed themselves as an anti-US anti-iMusic or anti-Pandora or whatever alternative. They conquered because their product was solid. Nobody cared where it's HQ was.

NoboruWataya 5mo ago
It seems clear to me that a company that pushes being EU-based and minimising US dependencies as a selling point doesn't really want to "conquer the globe". As a consumer it is even less important to me. Indeed, if your search provider has conquered the globe, then it's probably time to find a new one because the next logical step is turn the screws on users to maximise profits.
BrunoBernardino 5mo ago
Great point! In our comparison page with Kagi it should become clearer that’s not our only product differentiator. If you’re happy with Kagi, that’s great, it’s a very good product! We don’t want world domination. :)

Kudos 5mo ago

Country filters for Luxembourg and Monaco, no country filter for Ireland?

BrunoBernardino 5mo ago
Oops! Sorry about that. I’ll add it later today.

KomoD 5mo ago

> EU servers, EU storage, EU payment processing, EU search providers (Marginalia, Mojeek, EUSP, etc.). > All servers and data are physically in the EU. All search providers are based in the EU. Payment processing is done in the EU. Mojeek and Serper aren't EU so that's just false. And I'm not sure all the providers only use EU servers so I don't like the claim for that reason too. > Try in: Google // DuckDuckGo // Ecosia I would remove this, I thought it was for changing what provider I was using, but no, it just sent me to Google. Then at the bottom of the page is where I found this, which I would prefer to have at the top. > Try with: Mojeek // EUSP // Linkup // Serper

BrunoBernardino 5mo ago
Dang. You’re right that Mojeek is UK, so Europe but not EU (at the moment, at least). I’ll need to add a remark for that. We already do it for Serper. Regarding the servers, our servers and data are in the EU. We can’t guarantee that for our providers, but they don’t access your data. As for Try in/Try with, good point, we can add a preference for swapping them.

localhoster 5mo ago

Yavos - Yet Another Vibe-cOded Service :)

Hackbraten 5mo ago

Have you considered adding Google’s SERP to your indexes (via some third-party provider)? Google’s index is by far the largest, and my impression is that a search engine is hardly useful unless it includes Google’s results.

BrunoBernardino 5mo ago
Thanks! Serper is UK-based and does that. We provide them as an option, but not by default.

s_dev 5mo ago

Does Uruky pay Yandex like Kagi does? It's relevant to those of us boycotting Russian products and influence due to the Ukraine war.

BrunoBernardino 5mo ago
The search providers we use are listed in our FAQ and all are Europe-based. Yandex is not included.

dethos 5mo ago

Interesting. I will definitely add it to my list of services to experiment with. I like the account/payment system, where you top up a random account number for a period. Instead of having to go with the whole process of creating and verifying an account with your data and then managing yet another recurring subscription. Congrats on the release, wish you the best.

BrunoBernardino 5mo ago
Thank you! Please reach out if you have any suggestions or run into any trouble!

blfr 5mo ago

Can I (and how to) use it as a search engine for Open WebUI?

BrunoBernardino 5mo ago
I don't know what that is and it seems like some AI dashboard. Uruky has an API so as long as you keep within the rate limits, you should be fine!

AndroTux 5mo ago

Okay, so in order to do a test search, I have to: 1) type in a query and hit enter because there's no search button. 2) click signup, even though I want to evaluate it before creating an account. 3) apparently now I'm signed up without having to enter any details - what's the point? Just create a new session as soon as I initiate the search. 4) so now I need to return to the homepage to trigger another search. 5) search again, enter again. Now I'm greeted by a captcha. 6) after solving the captcha, I now have to enter my search query a third time because it wasn't saved 7) search results! Guys.

BrunoBernardino 5mo ago
The goal was to always show you the account number before you went for anything, but your post made me realize you don't need that if you're just trying it out, so now you should only: 1) type in a query and hit enter 2) solve the captcha 3) search results!
- AndroTux 5mo ago
  Much better. Now just add a freaking search button. That's literally your primary function, and you're lacking the button to do the one thing. I understand as a power user you don't press on buttons, but from a UX perspective, if you don't trigger the search immediately upon typing, you need a button to start the search. Your current UI suggests: Type in a query, and then click on "images" because that's the next thing that you can click. (Also apparently my IP is now banned and now every page just reads "Forbidden," so I couldn't even buy your product if I tried. You really should invest a lot in UX if you want this to be a product that's used by anyone other than hardcore privacy nerds.)

Havoc 5mo ago

Nice - I like this. Especially that I can also use it via API too and that expectations are limits are set out clearly (which seem very generous). On limits - consider changing the short limit on this: > searches to 1 per second, 30 per minute, and 1800 per hour. to 5 per 5 seconds or 10/10. That still works out to one per second but users are less likely to accidentally hit it with two requests that have similar timing. Say one from me and one from API usage.

BrunoBernardino 5mo ago
Good point. We have never had any reports of customers legitimately hitting these, but we can probably loosen them up a bit.
- Havoc 5mo ago
  Yeah can't picture the average user getting anywhere near the 2nd and 3rd limit. They seem very high to me, but the short one I can definitely picture hitting easily under API scenarios. e.g. Suppose you have 3 subagents getting triggered that all want to do a search. >can probably loosen them up a bit. Definitely wouldn't move the top 2. That would just be hard to walk back later
- Havoc 5mo ago
  I see this has already been updated in the FAQ . Thanks!
  - BrunoBernardino 5mo ago
    Thank you for the suggestion!

yegle 5mo ago

Given that: - Google has been around for 20+ years, so the concept of search engine and the technologies behind it should have been well known. - Computing power and Internet speed has increased significantly and in many homes (at least outside of US) 1Gbps is norm. - Everyone is talking about Google deteriorating over the years and prefer the old Google. The old system from the 2000s should be dirt cheap to run with modern home hardware. - People's need for search engine is highly specific, you presumably would be interested in searching a small subset of the whole Internet. My question is: why haven't a local run search engine be a thing at least in the tech circles? It should be able to bootstrap with e.g. an hourly updated "top 100 websites in 50 categories" index file, and adapt to my daily queries to automatically update the index in the background, and iteratively improves the quality of the results. The rise of the local LLM users proves this model works.

BrunoBernardino 5mo ago
You should check out hister.org!
- yegle 5mo ago
  Oh it checked multiple boxes I have in mind, and the website UI is really great. I'll definitely give it a try.
randomtoast 5mo ago
kagi exists for quite some time
- yegle 5mo ago
  I was talking about a self-hosted search engine.

ent 5mo ago

Please don't try to machine translate to my local language. The translations are terrible or outright incorrect.

BrunoBernardino 5mo ago
Thanks for reporting this. We’ve used a professional translator for German (though we’ve added many new phrases that haven’t gone through that process since), and we speak Spanish and French (besides English and Portuguese). We use DeepL for the rest but it’s not perfect. What’s your language, and do you know of a professional to help us? If you’d rather reach out to us via email, it also works.

dtj1123 5mo ago

One of the things I like about Mullvad is that I can buy a voucher on Amazon, and use that to pay for the service with a high degree of anonymity. Is/will something like that be possible with Uruky?

BrunoBernardino 5mo ago
I’m not aware of this, can you clarify? Technically anyone can top up an account on Uruky and give it away to someone else.
- dtj1123 5mo ago
  Like so, hopefully the link works for you: https://amzn.eu/d/00yFWx41 https://amzn.eu/d/00yFWx41 The card arrives with a little lottery-ticket style scratchpad with I believe a 16 digit activation code underneath.
  - BrunoBernardino 5mo ago
    Thanks, I see. You're still giving Amazon and Mullvad your payment information, so unless you're purchasing it for someone else, I'm failing to see the advantage? Again, that's something you can already do with Uruky as well.
    - dtj1123 5mo ago
      The advantage is that it makes it impossible for Mullvad or Amazon to associate my online payment with the activation code on the card, and by extension my Mullvad account.
      - BrunoBernardino 5mo ago
        Gotcha. We're now building vouchers for ProxyStore to sell them.
john_strinlai 5mo ago

Yizahi 5mo ago

Hi. A have a user report, after visiting the site it automatically switched to a language based on my system locale, but not on my UI language or keyboard preference. So it is possible to visit the site from some guest PC for example where locale would be something unfamiliar. And the language selector is down at the very bottom (I guarantee that my mom for example would never find it) it is plain text and it is in the currently set language. So for example if I will get a page set to Chinese or Korean or Hindi, I wouldn't even understand where language selector is, it would be a set of glyphs unrecognizable to me. Or vice versa, let's say an Indian person would get a cyrillic text. I suggest moving language selector to the header of the page and adding some icons to it - flags or English encodings (EN/DE/CN/RU/IN): https://en.wikipedia.org/wiki/List_of_ISO_639_language_codes https://en.wikipedia.org/wiki/List_of_ISO_639_language_codes

BrunoBernardino 5mo ago
Thanks for reporting this. Do you happen to know what your browser sends in the `Accept-Language` header?
- Yizahi 5mo ago
  I think this is what is sent from my side: "name": "Accept-Language", "value": "en-US,uk-UA;q=0.9,uk;q=0.8,en;q=0.7" "name": "User-Agent", "value": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:151.0) Gecko/20100101 Firefox/151.0"
  - BrunoBernardino 5mo ago
    Thanks, I've pushed a fix for this!
    - Yizahi 5mo ago
      Works on my side now, thank you. ;)
      - 5mo ago

maxloh 5mo ago

Could you consider Delayed Open Source Publication (DOSP) [1]? With that, you make the source code open-source after a specific period of time (such as 5 years). This way, your customers still get the source code and can use it freely, but you don't have to worry about competition. By the time anyone could use it to compete with you, they would be using an ancient version of your software. The BUSL (Business Source License) by MariaDB is a battle-tested license designed for exactly this use case. [1]: https://opensource.org/delayed-open-source-publication https://opensource.org/delayed-open-source-publication

BrunoBernardino 5mo ago
That is an option, for sure, but it feels less honest to our customers to give them the code "so much later". With the PolyForm Shield, they can just use the _current_ version however they want for themselves/family, as long as they're not competing with us commercially. Or am I missing something else, here?
- maxloh 5mo ago
  Not really. That means you give your customers the code to use under your current restrictions, and after a specific period, they can use it under the open-source license of your choice as well.
  - BrunoBernardino 5mo ago
    I see. We'll look into considering doing both, then. PolyForm Shield for the current code, with BUSL to transition into AGPLv3 after 2 or 3 years. Do you think that would work well?
    - maxloh 5mo ago
      Yeah. That would be great!
      - BrunoBernardino 5mo ago

npodbielski 5mo ago

> You can use the f=json search query parameter to get JSON-formatted search results, as per the FAQ entry above. Nice so EU-based company with API available for my agent to use. Right now I am using lightpanda but this my be simpler. I will check it out. Thanks.

zuzululu 5mo ago

I just tried to do a search and it immediately asks me to create an account??? I'm just not seeing the connect here I understand you are trying to fight spam/abuse but seems like a tough uphill battle especially when VPNs solve the privacy issue entirely. I'm also not sure why this obsession with EU is so prevalent and misguided. EU is privacy regulated, not a guarantee of privacy. GDPR is better than what many places have, but hosting in UE doesn’t magically mean the data is safe, it doesnt mean 0 knowledge encryption, no insider access, no breaches, no subpoenas, no natsec carveouts, or uniformly enforced across borders.

BrunoBernardino 5mo ago
"Create an account" is simply generating a random account number, but I _just_ pushed a fix to make that "auto-happen" for you if you try to simply search — you'll be presented with an option to solve a captcha or pay for access, then get redirected back to the search after that. The "EU obsession" for us is mostly about trying to buy and stimulate "the local economy" as much as possible, but YMMV and you're right it doesn't mean any of those things.

frangonf 5mo ago

Happy customer here. Found you last month while looking for searxng and related alternatives for API search with clankers in a personal setting.

BrunoBernardino 5mo ago
Thank you for supporting us!

prism56 5mo ago

I've just cancelled Kagi. While it was great. I'm trying to stop supporting US companies unless they provide good value

Havoc 5mo ago

uhm, I'm suddenly completed locked out? Can't even get to the homepage in private browse anymore so must be IP level. Just says "Forbidden". Must have tripped some sort of safeguard while trying to figure out how to use the API via curl

Havoc 5mo ago
OK this is a bit ridiculous. The FAQ does not specify how to use the API beyond listing a json parameter. If you send a curl command with that parameter the first search just returns "Logged In", and run another search and then you're temporarily IP banned site wide. Having hair trigger sensitive security around API while not telling the user what the request should look like to not trigger that is just silly.
- Havoc 5mo ago
  OK bit more trial and error later I worked out a curl command that doesn't trip this. https://pastebin.com/BNRBfYFf https://pastebin.com/BNRBfYFf Still not entirely sure what about the original request trigger it. Think it was something around the Sec-Fetch-* headers that copying a request from browser via copy cURL included
  - BrunoBernardino 5mo ago
    Sorry! You can use the incognito token (from the incognito URL) in order to not have to send the cookie over. If you were trying too aggressively to search, you could've been rate-limited for a few minutes.
    - Havoc 5mo ago
      Good to know about the incognito toke - helpful. Yes, I saw the 3 auth try mentioned in the faq after this, that must have been what I was hitting with the experimenting. Anyway all good now

Marciplan 5mo ago

As a European-first kinda person, I think we're good with Kagi?

BrunoBernardino 5mo ago
Kagi is a great product if you're happy with them! Uruky is for people who want ad-free, private, and personalized search, and no AI, while stimulating the European economy. That's not everyone and that's fine!

mietek 5mo ago

My number one request: let me block specific sites from my search results.

1kurac 5mo ago
FYI, on reasonable search engines like altpower.app, you can exclude sites with `-site:domain.tld` keyword. See also: https://altpower.app/#Advanced%20Search%20Operator%20Syntax https://altpower.app/#Advanced%20Search%20Operator%20Syntax https://www.exploit-db.com/google-hacking-database https://www.exploit-db.com/google-hacking-database
BrunoBernardino 5mo ago
`-site:domain.tld` works for "non-permanent blocking", but in your settings you can add domains or TLDs to block from the search results.

senko 5mo ago

Kagi's only office/hub is in Belgrade, which may not be EU, but it's (literally) close enough. Employees are remote. Freediver (founder) is US based and Kagi is an US entity, so must comply with any warrants there. But I guess they could set up a Serbian subsidiary?

Melatonic 5mo ago

Have you considered pivoting or prioritizing different test markets or niches? For example someone brought up making your UI and whatnot easier to use for someones "grandparents". What if you included a "grandpa" mode (you dont need to call it that specifically) that simplies everything and makes it super straight forward to use. I can see technically inclined people buying a subscription for their aging families to help them avoid clicking on useless (or dangerous) crap. You could also do something similar to Kagi SmallWeb but specifically for EU based stuff. Currently the UI is not bad but for some reason throws off my BS filter. I think it might be the spacing and (maybe) lines between the results - it sort of makes it resemble the paid text ads inserted into Google Search?

manytimesaway 5mo ago

Nice project! It lacks two things that are very useful : website blocking, and !hashbangs like DDG.

BrunoBernardino 5mo ago
I'm not sure if this is trolling, but we offer both. Even custom bangs, and you can also use Kagi's Bangs: https://github.com/kagisearch/bangs https://github.com/kagisearch/bangs
- manytimesaway 5mo ago
  This was not trolling at all, I didn't see these features on the website. Amazing, thanks for the quick response!

iqandjoke 5mo ago

Tried searching its name uruky on Uruky, some NSFW results shows up which I did not expected...

nutifafa 5mo ago

theamk 5mo ago

It would be super cool if you chose 20-100 queries (both common and niche) and shown the sample search results for those. Those would be completely cached/pre-rendered, so the cost would be very low and there will be no potential for abuse. And yet the visitors could at least get some idea about the quality of your search engine.

dukisamardzic 5mo ago

Solid EU-first positioning — the "trust local infrastructure" angle is increasingly important here. Curious about the privacy model: Kagi famously refuses to log queries. What's your approach to search history retention?

BrunoBernardino 5mo ago
No logging or tracking as well.

expedited123 5mo ago

Consider adding favicons near the websites in search results. Way easier to distinguish them at a glance. --- Also: The EUSP index is surprisingly good, by the way. Bingish level from my experience...

plextoria 5mo ago

This intrigued me and I topped up an account. Is there no password? Anyone could "steal" my account number and kick me out?

BrunoBernardino 5mo ago
There is no password, just like Mullvad. You should "protect" your account number like you would a password, but also, you can just discard it and change to another one in a month, for example.
- plextoria 5mo ago
  Thanks for your answer! Makes me curious how do you defend against people simply bruteforcing or guessing the account numbers.
  - BrunoBernardino 5mo ago
    Rate limiting when it's invalid (and high entropy to generate new ones)! It's in the FAQ :)