Introducing HIDN: A Room Without Names — Anonymous Confessions Built on Zcash

Hey Zcash community :zebra:

I want to share a project I’ve been building: HIDN (hidnroom.com).

What it is

HIDN is an anonymous confession app. You can say the thing you’ve never told anyone, reply to others, or let people message you and nobody (not other users, and not HIDN itself) can tell it was you.

There is no account. No email, no phone number, no username, no password. When you join, your browser creates a private key that only you keep. Every post, reply and like carries a zero-knowledge proof that shows “a real member wrote this” without revealing which member.

It’s the same kind of proof Zcash uses to keep payments private applied to words.

How the anonymity works

  1. Your browser generates a random secret key (the only way back in is by saving it).
  2. Only a commitment (fingerprint) of that secret is sent to the server. It joins a Merkle tree.
  3. Every action (post, reply, like) includes a Groth16 proof that your commitment is in the tree. The secret and Merkle path never leave your browser.
  4. One-time nullifiers prevent replay and stop proofs from being reused.
  5. Nothing links your posts together. Cosmetic “ANON-…” names change every visit.
  6. Private messaging links are end-to-end encrypted (messages are sealed in the sender’s browser; only the recipient’s device can open them).
  7. Tips go wallet-to-wallet in shielded ZEC. HIDN never holds funds or spending keys.

You can further hide your network address by using Tor Browser (Standard security level). Tho will be adding up a relay to keep it hidden at default

What you can do

  • Post and reply anonymously (up to 500 characters, no links allowed)
  • Choose from topic rooms (General, Things I’ve never told anyone, Talk about your ex, Ragebait me, Wildest thing you’ve done, Student secrets, Crypto confessions, and seasonal rooms)
  • Create private messaging links anyone can use to write to you anonymously
  • Enable tips so people can support posts or your private link with shielded ZEC
  • Browse a 3D “lobby/museum” where real confessions hang on the walls
  • Share posts as image cards
  • Get quiet, encrypted notifications for likes/replies on your posts

30-day organic stats (as of late October 2026)

In the first month of meaningful usage:

  • 153 keys created
  • 94 confessions
  • 29 replies
  • 51 likes
  • 40 private-link messages
  • 16 strangers activated tips

All organic traffic to https://hidnroom.com. No paid promotion.

What HIDN stands for

Most apps force you to hand over an identity before you can speak. That friction means the things people most need to say often never get said.

HIDN flips that

It treats privacy of speech the same way Zcash treats privacy of value

The content is public (by design confessions are meant to be read), but the who stays hidden. No per-user tracking, no retention metrics that require identifying people, no third-party analytics.

It’s an experiment in applying Zcash’s zero-knowledge tools beyond money: to everyday human expression.

some confessions from hidn room

Links

The code is open

The proofs run in the browser

Tips stay shielded and peer-to-peer.

If you value privacy as a default rather than an opt-in afterthought, come try saying something you’ve never said out loud.

Would love feedback from the community especially on the ZK circuits, the tipping flow, or ideas for better Zcash integration.

Stay shielded.
— Guardian / HIDN

1 Like

… Why do you need Groth16 at all? Anonymous chat rooms have existed for decades since USENET was a thing, and if you want proof of work, better and more predictable alternatives like RandomX or Argon2id exist.

1 Like

this is incredibe bro,it shows how zcash privacy system works.
I also have a build somewhat similar i made recently called “zarcade” feel free to check it out more details in my repo

1 Like

i mean, i just used it cos it was one of the easiest for me to implement as its also zero knowledge too

i mean, i am excited to check out your recommendation to see if work also on hidn

RandomX and Argon2id are not ZK algorithms, and I was suggesting using them for PoW, not for replacing Groth16. My point was, WHY do you need a zk algorithm here? If membership is not gated, then what’s the point of proving membership?

In fact, a Schnorr signature implies a SNARG, so it IS a succinct argument. You don’t need to make it zk.

EDIT: You know what? Fair enough. You can achieve unlinkability via this. There are alternatives such as ring signatures which are infinitely faster, but I mean, it’s a grant application, I’m sure the committee will handle it.

EDIT: Check my reply below.

To clarify my earlier reply: I was not suggesting Argon2id replace Groth16. Instead, in an ungated system, what are you verifying against? The posts aren’t linked, no identifiers are retained. Inclusion in a “membership set” that everyone is a member of? What prevents someone from just minting infinite Groth16 proofs? And then you’re just using end-to-end encryption, which is standard asymmetric cryptography, and don’t involve ZKPs at all. In fact, a Schnorr signature implies a SNARG!

Clearly, the goal here is to prevent someone from infinitely posting messages rather than actually proving anything, and a Argon2id challenge achieves that perfectly per-message while maintaining unlinkability, and it is infinitely harder to DDOS via. The secret is still maintained, but only used to set up the E2EE session.

Thanks for the questions

The membership set is intentionally open. Anyone can generate a secret and enrol its commitment into the Merkle tree. There is no gated allow-list.

The Groth16 proof is not there for traditional access control (“only approved people may post”). It is there to enable authenticated but fully unlinkable actions:

  1. You prove “I know a secret whose commitment is in the current Merkle tree” without revealing which leaf.
  2. You also reveal a nullifier derived from that secret + a scope (the specific post or action).
  3. The server only accepts the proof if that nullifier has never been seen before for that scope.

This gives two properties at once:

  • Unlinkability — no one (including the server) can tell whether two posts, replies, or likes came from the same member.
  • Scoped one-time use — the same membership cannot be reused for the same action (e.g. you can’t like the same post twice, and a captured proof can’t be replayed).

Without this the alternatives are weaker:

  • Completely free anonymous posts → no good cryptographic way to limit abuse while keeping unlinkability.
  • Ordinary signatures → actions become linkable by public key.
  • Pure Argon2id / PoW → useful for raising the cost of spam, but it doesn’t give a growing anonymity set or clean scoped nullifiers.

Ring signatures are a legitimate alternative and can be faster in some cases. We chose a fixed-depth Merkle tree + Groth16 because:

  • Proof size and verification cost stay constant as the set grows (up to ~1M members).
  • It follows the same nullifier pattern used in Zcash.
  • The circuit is simple and the tooling was already familiar.

You’re right that a Schnorr signature is itself a SNARG. The zero-knowledge property is what hides which member is acting (that is the part we need.)

Server-side rate limits still exist on top of the crypto layer. The ZK system’s job is the unlinkability guarantee, not to be the only anti-spam mechanism.

Happy to discuss the circuit, nullifier scopes, or the trade-offs versus ring signatures.

i mean,

i am also happy to explore other ways your suggestions can work

the end goal is privacy and ensuring that people get to shield their thoughts privately

I- Even though the “pattern” of how that reply is written is… familiar, I’ll still answer. You literally wrote up all that and said absolutely nothing.

An Argon2id hash IS a nullifier, and the used hash set IS the anonymity set. It’s simple logical derivation. Unless there is some other identifier of users that the Groth16 proof proves, but then it isn’t unlinkable!

You are literally saying that “I won’t use Y because even though Y has all the properties I need, it doesn’t have the same name for the properties!”

An Argon2id hash size is constant and can be verified in O(1) time in a hash tree, for nearly infinite members via chunked trees.

so what you’re saying is the Argon2id hash is better option and use of tech then the groth?

cos as opposed to the groth that’d have the zk, it doesnt have all the properties

in this case scenario?

i mean, if argon2id hash can work and can be verified, unlinkable and also keep thr privacy

i am happy to explore using it

… Yes? Obviously? Not just in this case scenario, any scenario where you don’t actually need to prove anything.

Plus, even if you DID need to prove ownership of an account, you can use one of a gazillion asymmetric key DSA schemes which imply SNARGs/SNARKs, instead of using a general purpose SNARK which is practically infinitely more inefficient.

amazing

happy to explore using it in that case.. i am not really a technical person but happy to get it setup, and you could also recommend ways i could get to use it or try it

main focus for me is preserving the privacy for the users and ensuring its unlinkable

Yet you kept giving me AI-generated responses back while pretending it was you. That’s material misrepresentation.

Using an AI isn’t bad, pretending you aren’t is. You should first learn cryptography, ideally do a cryptography course (not a crash course, a proper, university level course, ideally Master’s level) before building things like these, because a potentially flawed or false security claim is infinitely worse than no claim at all.

appreciate your feedback and yea i’d look into it

but never did i claim or gave a false security claim nor did i pretend

in all my responses i was open to see and learn how i can get it done and explore that

that shows my interest to also explore other options

so i never gave a security false claim, and from your replies you gave a thought on using Argon2id… which is a better and more predictable alternatives like you mentioned

and like i said, i am more focused on ensuring HIDN is secure, private and unlinkable

And i am open to checking argon like you mentioned

all love

happy to also connect with you, maybe on discord, telegram signal, x or something

so i can share learnings and discoveries with you when i explore argon

Sure. I’m on Signal as violetb.06.

thank you

trying to get my signal restored, will dm as soon as i resolve it