PARSEC is the reference sovereign, client-side, per-connection-keypair wallet under CP2048-QR — and the project where the cypherpunk lineage first earns Tier-Q (quantum-native), on Algorand. Tier-Q carries forward into cypherpunk4096 (2¹², the strict superset that LUV/DeltaVerse now build to): cp4096 commitment #5 is quantum compliance — signatures handled as bytes, never (v,r,s), scheme-migratable behind a timelock — and PARSEC is where that commitment is realized in a live wallet. This is the Algorand track, separate from the ETHGlobal/EVM submission.

Grounded in Algorand's Post-Quantum Cryptography Roadmap (read 2026-08-20). This file tracks that roadmap; re-check it before shipping any PQ claim.

Account-level post-quantum is arriving as a native, standards-tracked capability — not a bolt-on. The pieces that matter for a wallet:

• Native Falcon-1024 accounts — Q3 2026. Supersedes the LogicSig-only Falcon pattern (the 2025 first-PQ-transaction proof). Native address derivation is SHA512_256(domain ‖ scheme[2] ‖ explicit_salt[1] ‖ pk), preserving the 32-byte address / 58-char format, and Falcon-1024 is derivable from the standard 25-word seed — so a PQ account looks and seeds like a normal Algorand account. Legacy SDKs, Pera Wallet, and AlgoKit will support it.
• Cryptographic agility — Q3 2026. Network-level support for multiple concurrent signature schemes (Ed25519 alongside Falcon; FN-DSA, ML-DSA, others under the agility framework), with no further structural overhaul to add a scheme. This is the substrate version of cp4096 #5.
• Hybrid accounts. "Any combination of keys" — merge an ECC account with a lattice one for defense against both classical and quantum risk. This is PARSEC's cutover path (run ed25519 + Falcon together).
• Native PQ multisig — end 2026. A generic policy layer over independently-verifiable signatures (weighted approvals, hybrid classical+PQ signers), not scheme-specific threshold crypto.
• Falcon-512 — end 2026. Smaller keys/sigs for cheaper accounts.
• State Proofs (Falcon) already give quantum-secure finality / light-client proofs today.

Key/sig sizes (the concrete reason cp4096 forbids (v,r,s)):

A Falcon signature is 640–1280 bytes — it cannot fit the 65-byte (v,r,s) shape, which is exactly why cp4096 mandates signatures be carried as bytes. The roadmap is the empirical justification for that commitment.

State this plainly wherever PARSEC claims Tier-Q:

• Accounts/signatures: quantum-resistant is reachable now (Falcon LogicSig today, native Q3 2026).
• Consensus is still Ed25519 — voter signatures and the ECC VRF (cryptographic sortition) are not yet quantum-resistant. Algorand's PQ-consensus + VRF-replacement research paper is expected early 2027; PQ consensus messaging details late 2026. State Proofs cover finality proofs, not live voting.
• So PARSEC on the Algorand track is Tier-Q at the account/signature layer, on a network whose consensus quantum-resistance is on the roadmap but not shipped. Label it that precisely — never "quantum-proof wallet" full stop.

1. Falcon (FN-DSA) keypair per connection, client-side, open-source — use FALCON-DET1024 (deterministic, integer-only keygen, no floating point; the variant Algorand's Trezor Safe 5 PoC ships: ~2.2 s keygen, ~0.69 s sign). Floating-point Falcon is an implementation hazard; the wallet uses the integer-only path.
2. Target NATIVE Falcon-1024 accounts (Q3 2026), derived from the standard 25-word seed via the native derivation above — not a bespoke LogicSig. PARSEC inherits this the same way it inherits Algorand support generally: from Pera Wallet's code (Pera is a named roadmap supporter), matching PARSEC's canonical-wallet-as-example compatibility model (Pera→Algorand, Phantom→Solana, MetaMask→EVM). Keep the LogicSig path as the pre-Q3-2026 fallback.
3. Hybrid through the cutover — an ed25519 + Falcon hybrid account so a connection is safe against both adversaries during transition; drop ed25519 only when native PQ is battle-tested.
4. Anchor identity/session commitments to Algorand State Proofs (quantum-secure finality).
5. PQ multisig for the OVERLORD / DAIO 2-of-3 surfaces via Algorand's generic-policy multisig (end 2026) — weighted, hybrid classical+PQ signers.
6. Vault stores bytes, not (v,r,s) — the bankon_vault keyring already holds raw key material across architectures (ed25519 · secp256k1 · RSA-4096); add Falcon as another bytes scheme, no format assumption that a signature is 65 bytes.

The roadmap is a point-for-point substrate realization of cp4096's quantum commitment (#5), and PARSEC is the wallet that consumes it:

• bytes, never (v,r,s) ↔ Falcon sigs are 640–1280 B. The vault, the signer interface, and any on-chain verifier PARSEC touches must treat a signature as opaque bytes.
• Scheme-migratable behind a timelock ↔ Algorand's cryptographic-agility framework (concurrent schemes, add-without-overhaul). cp4096's timelocked scheme swap is the contract-layer mirror of Algorand's network-layer agility; the bankonvault lockers' i_signature_verifier crypto-agility seam is the same idea one layer down.
• Hybrid classical+PQ ↔ Algorand hybrid accounts + hybrid multisig — the honest migration posture cp4096 asks for (never claim PQ before the PQ scheme is live and battle-tested).
• Determinism / no approximation ↔ FALCON-DET1024 integer-only keygen (no floating-point nondeterminism), which is also what makes a hardware-wallet and cross-device implementation reproducible.

PARSEC is itself sovereign and (on the Algorand track) heading to quantum-resistant accounts, modularly attaching to external, non-sovereign wallets while each connection holds its own client-side keypair — the reference shape of CP2048-QR §3, carried into cp4096. See RAILS.md in the standard.