The Hidden Wiki began as a community-edited index of onion sites, back when there was no other way to find anything on the Tor network. The name is now carried by dozens of unrelated sites, almost none connected to the original, and a good many existing to take money off people who trusted it.
This is a history, not a directory. There are no addresses here and no links to any site of that kind, deliberately. By the end it should be obvious why handing you a list would be the least useful thing anyone could do.
What it was
In the late 2000s the Tor network had no search. Onion addresses were sixteen characters of noise that couldn’t be guessed, inferred or indexed. Without an address you couldn’t reach a site, and there was nowhere to find one.
The Hidden Wiki solved that the obvious way: a wiki, running as an onion service, editable by anyone, listing addresses by category. It indexed a genuine mixture — email providers, forums, political material, file hosting, and criminal markets sitting in the same list as all of it.
For a while it worked roughly as intended. People who cared about it maintained it, entries got corrected, dead links got pruned.
How it came apart
Three things pulled at it, and they fed each other.
Open editing. A wiki anyone can edit is a wiki anyone can vandalise. Entries got swapped for links to other things. Reverting turned into a full-time job nobody was paid to do.
Takedowns. When hosting providers were seized — most consequentially in 2013, when a single host carrying a large share of onion sites went offline — the wiki and much of what it indexed disappeared together. Each time, copies surfaced claiming to be the continuation.
No way to prove anything. This is the structural problem, and it never got solved. There was never an authority who could say which copy was real. Nobody owned the name. No maintainer could be pointed to as legitimate. Every fork had exactly as much claim to the title as every other, which is to say none whatsoever.
By the middle of the 2010s there was no such thing as the Hidden Wiki. There were many, contradicting each other.
What finished it
The retirement of v2 onion addresses in 2021 broke every entry in every copy at once. Whatever fragments of the original were still limping along stopped being usable overnight, along with everything they pointed at.
What filled the gap wasn’t a community rebuilding an index. It was a large number of sites, cheaply generated, carrying the name because the name still had search traffic attached to it.
Why the copies are dangerous
The mechanics are worth following, because they generalise well past this one name.
A directory publishing onion addresses sits in an unusually powerful position, for a reason specific to how these addresses work. Fifty-six characters of apparent randomness can’t be verified by eye. Nobody memorises them. Nobody spots when one changes. If a directory lists an address differing from the real one somewhere in the middle, essentially no visitor will catch it.
So the attack is simple. Generate an address whose first and last characters match a real one, since that’s the part people glance at. Stand up a copy of the target site at that address. Get it listed. Wait.
Everything the visitor types goes to the operator. Every payment goes to the operator. And because the two sites look identical, the visitor may never work out what happened, or when.
A copy of a directory doesn’t need to look suspicious. It needs to look helpful, which is a much easier brief.
The detail that keeps surprising people
Having watched this cycle go round several times: the copies get better at appearing legitimate, not worse.
Early ones were obviously slapped together. Current ones carry safety warnings, editorial notes about which entries have been verified, last-checked dates, and sensible advice about avoiding scams. Some contain genuinely accurate general information.
That isn’t a contradiction. Trustworthy presentation is cheap, and it’s exactly what makes the poisoned entries work. A directory warning you about phishing while listing a phishing address is more effective than one doing neither.
Judging these sites by whether they seem responsible stopped working years ago.
So what do you do instead
The honest answer is that the thing the Hidden Wiki existed to provide can’t be provided safely by anyone. Including this site.
An address is only trustworthy when it comes from whoever it belongs to. A newspaper publishing its own onion address on its own website, over an encrypted connection, is a chain you can follow and check. A list compiled by a stranger isn’t, no matter who compiled it, because there’s no entry on it you can independently verify.
The Tor Project does this properly, for what it’s worth — it publishes its own onion addresses on its own infrastructure, which is what a legitimate address source looks like. Note that it’s publishing its own and nobody else’s. That’s the whole distinction.
It makes the network harder to explore than the ordinary web. That’s a genuine cost and nobody has found a way around it. Directories don’t solve the problem. They move it somewhere you can see it less clearly.
Next
What an onion service is explains why these addresses can’t be checked by eye. What Tor hides and what it doesn’t covers why the network protects your route but never your judgement.