AntiQua has undergone two full external audit rounds:
NightShift Audit — 47 findings identified and remediated.
Draikoon Audit — 11 findings, all CRITICAL and HIGH resolved.
Production security decisions include: hard-abort on plaintext key handling, mempool RBF lockout, DNS eclipse fallback seeders, and mnemonic export PIN protection.
Bug bounty program incoming. Details to follow.
AntiQua ships with seven built-in smart contract templates, all tested (257/257 passing) :
Custom contracts beyond these templates are on the roadmap. Discussion welcome.
AntiQua spans three repositories:
NodeCore (C++) — The full node. P2P networking, UTXO model, mempool, smart contract VM, RandomX PoW, ACRP chat layer.
Desktop Wallet (C# WPF) — Full-featured Windows wallet with transaction history, smart contract interaction, and encrypted ACRP messaging.
Android Wallet (Kotlin) — Mobile node + wallet. Runs a full AntiQua node on your phone.
All three are developed simultaneously. No forks. Every line written from scratch.
Bitcoin uses BIP-39 mnemonics backed by secp256k1 keys — fundamentally incompatible with post-quantum cryptography. AntiQua implements AIP-39, a custom mnemonic standard designed around ML-DSA-87 key material.
Your seed phrase on AntiQua is quantum-resistant from word one.
Almost every blockchain in use today secures your coins with elliptic-curve signatures (ECDSA or EdDSA). They are fast, compact and safe against every computer we have today. Against a large quantum computer running Shor's algorithm, they are not safe at all.
AntiQua was designed from day one on the assumption that this threat is not hypothetical – it's a matter of when. Here is what that means in practice. 
What a quantum computer actually breaksNot everything in a blockchain is equally at risk. The real problem is signatures, not hashing:
| Building block | Used for | Quantum attack | Impact |
|---|---|---|---|
| ECDSA / EdDSA | Signing transactions | Shor's algorithm |  Broken – the private key can be computed from the public key |
| RSA, ECDH | Encryption, key exchange | Shor's algorithm | Broken |
| SHA-256, SHA-3 | Hashing, addresses, mining | Grover's algorithm |  Weakened – effective security roughly halved (256 → ~128 bit), still considered safe |
| AES-256 | Symmetric encryption | Grover's algorithm | Weakened the same way, still considered safe |
In plain words: whoever controls a large enough quantum computer can derive your private key from your public key – and sign transactions as if they were you.
Why "not yet" is not a good excuse1. Public keys live on-chain forever.
On most blockchains, your public key becomes visible the moment you spend from an address – and it stays in the chain history permanently. Every coin behind an exposed public key is a target the day a capable quantum computer exists. That includes millions of coins already sitting in such addresses today.
2. Migrating a live blockchain takes years.
Switching the signature scheme of an existing network needs consensus, new wallets, exchange support – and every single holder moving their funds. Coins whose owners lost their keys or stopped paying attention can never be moved to safety.
3. The standards are already here.
In August 2024, NIST published the first post-quantum standards: FIPS 203 (ML-KEM) and FIPS 204 (ML-DSA). NIST's transition plan proposes to deprecate quantum-vulnerable algorithms like RSA and elliptic-curve cryptography after 2030 and disallow them after 2035. The rest of the industry is starting its migration now.
4. Attackers can wait.
"Harvest now, decrypt later" is not science fiction: data and public keys collected today can be attacked the day the hardware is ready.
How AntiQua handles itAntiQua doesn't bolt post-quantum cryptography onto an old design. It is the design:
| Layer | AntiQua uses | Standard |
|---|---|---|
 Signatures |
ML-DSA-87 | FIPS 204 · NIST security category 5 |
 Key encapsulation |
ML-KEM-1024 | FIPS 203 · NIST security category 5 |
 Hashing |
SHAKE-256 | FIPS 202 (SHA-3 family) |
 Seed phrases |
AIP-39 | our own post-quantum mnemonic standard |
 Proof of Work |
RandomX | CPU-friendly, ASIC-resistant mining |
NIST security category 5 is the highest level NIST defines: breaking it must be at least as hard as brute-forcing a key for AES-256.
And because post-quantum signatures are there from the genesis block, there is no legacy signature scheme to migrate away from – no future hard fork just to become quantum-safe.
The honest part: where the cost is – and how we contain itPost-quantum security isn't free. The keys and signatures are much larger than what Bitcoin or Ethereum use:
| Public key | Signature | |
|---|---|---|
| ECDSA (secp256k1) | 33 bytes | ~72 bytes |
| ML-DSA-87 | 2,592 bytes | 4,627 bytes |
| AntiQua combined key (ML-DSA-87 + ML-KEM-1024) | 4,160 bytes | – |
A naive design would store all of that forever, on every node. AntiQua doesn't. Signatures are expensive once – on the wire and when a block is first accepted. After that, the network works with hashes, compact commits and checkpoints.
Verify once, then commit compactWhen a block is connected, every transaction is verified in full – including its ML-DSA signature. What gets written to the ledger afterwards is a compact record: the 32-byte transaction hash, inputs, outputs and amounts – without signature and public key. The Merkle root and the block header hash bind that record to the verified original. Full and light nodes therefore store the proof of validation, not the proof itself.
Commitments instead of bodies in gossipBlock announcements carry transaction commitments – again 32 bytes per transaction. The full block with all signatures only goes to a subset of verifying nodes, or to whoever explicitly requests it.
Checkpoints instead of re-checking history Why this stays secureThe check always happens before anything is discarded: a compact record is only valid because the full transaction was accepted first. Archive nodes keep the complete bodies, so anyone who needs the original – for an initial sync or a wallet – can fetch it and verify it again against the committed hash.
In short: post-quantum signatures cost bandwidth and CPU once. Storage stays lean. And even if it didn't: storage gets cheaper every year – a stolen private key never gets un-stolen.
Let's talkQuestions, doubts, counter-arguments? That's exactly what this category is for. Happy to go deep on any part of the stack – ask away
Want more? Read how AIP-39 keeps your seed phrase quantum-resistant, or check our Security Audit History.
— Blythex
As of March 31, 2026, the AntiQua testnet went live with four simultaneous nodes syncing correctly across Hetzner, home node, and two Android devices.
The full stack is operational:
Stay tuned.
Welcome to the official AntiQua community forum 
This is the home base for everyone building, running or simply curious about AntiQua – a blockchain designed from the ground up for the post-quantum era. Whether you're a developer, a miner, a cryptography nerd or just found us today: you're in the right place.
AntiQua at a glance| Cryptography |
NIST Level 5 – ML-DSA-87 signatures, ML-KEM-1024 key exchange, SHAKE-256 hashing |
 Consensus |
RandomX Proof of Work |
 Supply |
56,000,000 AQA – hard cap, 75 % mined + 7 % node rewards |
 Launch |
Fair launch · clean genesis block · no VC allocation · funds multi-sig or vested |
 Codebase |
Built from scratch – no forks, no shortcuts |
 Status |
Testnet live since March 31, 2026 |
New here? Start reading
Why Post-Quantum? The Threat is Real – why today's blockchains are at risk and what AntiQua does differently AIP-39 – Our Post-Quantum Mnemonic Standard – how your seed phrase stays quantum-resistant AntiQua Testnet is Live – where the network stands today
Repository Overview – NodeCore, Desktop Wallet and Android Wallet
Tokenomics – the 56M AQA supply, the Relay Reward Lottery, the four funds and the rules that bind them Security Audit History – our external audits and what came out of them
Find your way around
Announcements – official news from the core team
Technology – post-quantum cryptography, consensus and protocol internals
Development – NodeCore, wallets, code and bug reports · follow the NodeCore Changelog Mining & Nodes – running a node and mining AQA with RandomX
Support – stuck? Read How to Get Help Fast, then ask away
Economics & ICO – tokenomics, supply and distribution Security – audits, hardening and responsible disclosure
Community – introductions and everything else
Your first steps
Say hello in Introduce Yourself – tell us who you are and what brought you here
Watch the NodeCore Changelog (bell icon → Watching) to get notified about every core update
Write in your own language – every post has a translate button next to Reply Have a quick look at our Community guidelines – short and to the point
Stay safe: The AntiQua team will never ask for your recovery phrase, your private keys or remote access to your computer. Anyone who does is a scammer – please report them.
We're building AntiQua in the open, and this forum is where it happens. Ask questions, challenge our decisions, report bugs, share ideas – we read everything.
Glad to have you here. 
— Blythex, Lead Developer
Update — Telegram notifications are liveSeptember 2026 · a new plugin joins the list above
You can now get your forum notifications on Telegram — if you want to. Nothing changes for anyone who doesn't: email and the bell keep working exactly as before, and this is off until you switch it on yourself.
Our bot is @AntiQua_Blockchain_Bot.
Connecting takes about ten secondsThat's it. And notice what you didn't do: you never looked up a chat ID and never typed one anywhere.
That's deliberate, and it's the most important design decision in this plugin. If there were a field for your Telegram ID, someone could type in yours. Their forum notifications would then land on your phone — and since notification texts contain post content, whoever did it could choose the text. The forum would become a delivery channel for messages from a bot you trust. That is exactly how a good phishing message gets made.
So instead: you get a one-time link, you press Start, and Telegram itself tells the bot who you are. Entering someone else's ID isn't hard in this design — it's impossible, because the ID never passes through anyone's hands. The link works once, expires after ten minutes, and only its hash is stored in our database, never the link itself.
What will never arrive on TelegramPassword resets. Email confirmations. Two-factor codes. Those stay on email, permanently, with no setting to change it.
The reason is simple: if a password reset could be delivered to a messenger, then whoever controls your Telegram controls your forum account — and the email confirmation that your account is built on would be bypassed entirely. Convenience isn't worth that trade, so we didn't build the switch.
You choose what arrivesEach kind of notification has its own toggle:
On top of that there's one master switch for Telegram delivery, and Disconnect removes everything. You can also send /stop to the bot — same effect, from wherever you happen to be.
It speaks your languageThe bot picks your language in this order: whatever you select in the Telegram section, otherwise your forum language, otherwise the language your browser asked for when you connected, otherwise the forum default. All 14 forum languages are supported.
That third step matters for people who never set a language in the forum — the browser already knows, so the bot simply uses it instead of defaulting to English.
Small things that keep it pleasant What we store, and what we don'tYour chat ID, your display name, when you connected and which kinds you chose. That's all.
It disappears when you disconnect, when you send /stop, and when you delete your account. In server logs and in the admin overview the chat ID only ever appears shortened (****1570) — it's personal data, and it gets treated that way.
Tested the way we'd want it testedLike the plugins above, this one ships with its own test suite — 59 tests, including the attacks we'd expect someone to try:
All of them fail, which is the point.
Questions, or something behaving oddly? Reply here. If you'd rather not use Telegram at all, you genuinely don't have to — nothing about the forum requires it.
I think that by the first quarter of 2027, the first nodes will be running on the testnet by users.
Когда можно будет присоединиться к сети?
