What is Arrekta
The quantum-safe settlement layer for issued assets. Arrekta is a public proof-of-work, multi-asset UTXO chain where stablecoins, tokenised deposits and other real-world assets are issued and settled with NIST post-quantum cryptography today.
- Post-quantum by default. ML-DSA (FIPS 204) signatures, a hybrid X25519 + ML-KEM-768 encrypted P2P transport, and SHA3-256 hashing everywhere. See Post-quantum cryptography.
- Native issued assets. Issue, mint and burn assets without contracts. Optional issuer controls (freeze, clawback) are declared at issuance, public, and enforced by consensus. See UTXO & native assets.
- Bitcoin-merged-mined. Blocks prove their work with an AuxPoW in Namecoin's format, so Bitcoin pools can secure Arrekta at no extra cost. See Merged mining.
- Signed checkpoints for finality. A public, rotatable post-quantum signer group makes blocks final; no reorganisation crosses a final checkpoint. See Checkpoints & finality.
- Spending policies, not a VM. Multisig, timelocks, hashlocks, HTLCs, vaults and vesting come from one small audited policy language. See Locks & policy locks.
- Two independent implementations. A Rust production node and a Python reference node agree byte for byte on shared conformance vectors.
Run a regtest node, mine blocks and send a payment with the SDK.
Assets, locks, PST, merged mining, finality, fees, light clients.
TypeScript and Python clients over the node REST API.
The normative specification, rendered from core/PROTOCOL.md.
Why a settlement chain
Arrekta's benchmark for "solid" is Bitcoin and Liquid: a simple, conservative base layer that does a few things and never breaks. That means:
- Nobody can rewrite history cheaply (merged mining plus signed checkpoints).
- The rules never change by accident (normative spec, two implementations, shared vectors).
- The network stays up under attack (P2P hardening, DoS limits).
- Miners keep being paid enough to secure it, forever (a permanent 0.6 ARRK tail emission).
- Operators can run it with confidence (monitoring, signed reproducible releases, incident process).
Some features of application platforms are deliberately not on layer 1:
| Feature | Why Arrekta doesn't need it |
|---|---|
| General smart contracts (EVM/SVM) | Issuance, restrictions, freeze and clawback, multisig, escrow (HTLC), vaults, vesting and atomic swaps (PST) are native. A VM is a large attack surface and weakens the "assets behave predictably" promise. |
| Proof of stake | PoW keeps fair issuance and no validator set to capture; finality comes from signed checkpoints without stake. |
| Blob data availability, rollups | Settlement volume doesn't need them yet. |
The door stays open: reserved lock kinds and version-bits deployments let any of these arrive later as a soft fork.
How it compares
| Bitcoin | Ethereum | Solana | Arrekta | |
|---|---|---|---|---|
| Consensus | PoW | PoS | PoS | PoW (merged with Bitcoin) + ASERT, signed checkpoints |
| Post-quantum | proposals (BIP-360) | roadmap target | none | ML-DSA, ML-KEM, SHA-3 today |
| Native assets with issuer controls | no | contracts | token program | yes, enforced by consensus |
| L1 throughput | ~7 TPS | ~15–30 TPS | thousands | ~12 TPS (≈35 with Falcon, once FIPS 206 is final) |
| Independent implementations | several | several | 2 | 2 |
The full comparison, with sources, is in Layer architecture and L1 gap analysis.
The stack
| Layer | What | Status |
|---|---|---|
| L3 products | indexer (Esplora-compatible), gateway (API keys, quotas, PQ-signed webhooks), wallet service, issuer service, live streams, SDKs | built; SDKs in progress |
| L1 ledger | UTXO multi-asset, issuer controls, policy locks, PST, compact filters, soft-fork path, fee bumping, state commitment | built |
| Consensus | merged-mined PoW + ASERT, signed checkpoints, tail emission | built |
| L0 network | PQ transport, headers-first sync, compact blocks, eviction and ADDR limits | built |