Shutting Down Our Search Proxy Leta
Postedabout 2 months agoActiveabout 2 months ago
mullvad.netTechstory
calmneutral
Debate
20/100
PrivacySearch EnginesGoogle
Key topics
Privacy
Search Engines
Google
Mullvad is shutting down their search proxy Leta, sparking discussion about the challenges of scraping Google and the implications for private search.
Snapshot generated from the HN discussion
Discussion Activity
Light discussionFirst comment
1h
Peak period
1
1-2h
Avg / period
1
Key moments
- 01Story posted
Nov 6, 2025 at 8:44 AM EST
about 2 months ago
Step 01 - 02First comment
Nov 6, 2025 at 9:53 AM EST
1h after posting
Step 02 - 03Peak activity
1 comments in 1-2h
Hottest window of the conversation
Step 03 - 04Latest activity
Nov 6, 2025 at 2:15 PM EST
about 2 months ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
ID: 45835161Type: storyLast synced: 11/17/2025, 7:55:10 AM
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
The main problem is that Google desperately wants you to run their JavaScript. They implemented a JS requirement a few months ago, but you could still pretend to be the Lynx web browser and then not be served any JS. Then a little bit ago they removed Lynx support, so that stopped working. There were a few other things you could pretend to be that still worked though, like an old iPod I believe. Then those were blocked as well, but currently there is still at least one user agent string that works.
Some of this I learned about from a proxy called 4get.ca. The guy who runs it is clever and has the energy (for now) to keep the Google scraper working. The caveat though is that he has a somewhat crude and demented sense of humor.
Even if all the user agents are blocked, I worked with him a bit on an alternative bypass which is a bit slow, and is based on my article in the most recent issue of 2600 Magazine (Autumn 2025 if you want to read it). If need be I can probably make it faster. He also has a bunch of Google CSE api keys and makes that available as a user selectable option.
It isn’t a meta-search engine because it doesn’t mix results together from different sources and re-rank them, it’s just a proxy, but as a proxy it’s pretty good.