Security and Transparency.
A currency in development.

01 / ARCHITECTURE

Validate independently.

The current Go lab has signed account transfers, two validating primary nodes, an additional test node and a read-only explorer. Its JSON persistence and simple HTTP peer exchange are unsuitable for a public financial network. A separate, undeployed Core fork explores a more mature privacy architecture. Bounded two-host trials relayed mined blocks and wallet transfers, then split an outbound-only edge onto short competing branches. A later trial confirmed a transfer on the losing edge branch, returned it to pending after reorganization, and reconfirmed it in the intermediate peer's next block with all three nodes agreeing. A separate two-host test showed that the edge retried after a link cut, but a stale listener socket blocked reconnection until that socket was cleared; the next block then reached both nodes. The inherited minimum transaction rebroadcast delay is five minutes, longer than the earlier 90-second pool check. A follow-up two-host trial left the stale socket untouched: its timed sync eventually closed the socket, the edge held a recovered peer without daemon restart, and both nodes accepted the next block. A separate private-interface Core test also connected from an operator-supplied peer, saved that peer, reconnected without the startup peer option, caught up offline blocks and relayed a new block back. Both ordinary and priority peer options passed. A follow-on three-node test found that private addresses learned from peers were rejected even with the local-peer option; the experimental Core now admits them only under explicit local-peer opt-in. After its bootstrap peer stopped, the third node selected a learned alternative and synchronized new blocks. A subsequent two-host trial confirmed the outbound-only node could learn a distinct alternate, switch to it when bootstrap stopped, restart without a peer argument, catch up from its saved peer cache, and relay a new block back to the same height-7 tip. The experimental peer manager now filters advertised public endpoints with no usable IP address or peer port, including saved entries; this is a narrow input check, not hostile-discovery proof. These remain single-operator private-network tests; public discovery, malicious peers, real NAT and scale remain unverified. Wider recovery timing, real NAT and timed multi-hop relay remain open; these trials use provisional fixed-difficulty consensus. Privacy strength and final fee behavior remain unreviewed. One integrated Core chain with final consensus rules, independent peer discovery and measured scale remains the next milestone. No Core network is production-ready. Independent operators must be able to validate without founder infrastructure before launch.

A newer two-host Core trial put an outbound-only node behind Docker bridge NAT with no published port. It accepted a mined block, relayed one back, and caught up after a brief pause. Ordinary home-router NAT, independent bootstrap and public scale remain unverified.

02 / MONETARY POLICY

Make every unit accountable.

The v4 lab fixes a historical issuance ceiling of 25 million PNPT: one million in genesis and 24 million allocated to mining across 960 height-based years. Its first-year schedule is 25,000 PNPT per 262,800 blocks. Difficulty targets a two-minute average, but calendar-time issuance varies with actual block arrivals. The proposed founder allocation would lock for two height-years and release over the next three; that vesting is not enforced by the live lab. A separate, undeployed Core issuance prototype now validates an eight-decimal height-based mining subsidy on a fresh chain. Offline validators accepted its first scheduled RandomX reward and rejected a one-atom overpay and insufficient work; wallet transfer and recovery, restart, and three-node relay passed. Two bounded reorganization tests also passed: a detached payment returned to pending, while a conflicting losing spend was rejected and stayed absent after restart. The earlier issuance branch has zero founder coins. A separate, undeployed funded-genesis lab creates exactly one million PNPT at genesis and mined the first scheduled reward in a disposable two-node test. Its publicly known fixture key is insecure, and its single output locks all at once rather than gradually vesting. A bounded two-host Core trial relayed mined blocks both ways and recovered after one node restarted; a separate funded-chain wallet trial confirmed a transfer and seed recovery. Historical-cap enforcement and fee-bearing reorganization fixes passed bounded same-host tests at fixed difficulty. Separate founder-release and inventory models do not enforce vesting or change active wallets. Founder vesting, complete historical-cap validation, final fees and independent review remain open. The running Go lab is unchanged. The final fee market and burn rule are undecided.

03 / MINING

Participation needs honest controls.

The lab's solo miner pays a selected public wallet address and offers a 5–100% duty-cycle setting for one worker. That setting changes hash attempts, not the reward for a block, and is not a whole-computer CPU cap. Current SHA-256 work favors specialized hardware. The separate Core fork passed a deterministic difficulty-shock input test, but a 100-fold hash-rate loss still delays its first modeled block by about 3 hours 20 minutes. Raw timestamp stress inputs exposed a sharp trimming boundary. A short Core validator trial rejected blocks below the recent timestamp median or beyond the two-hour future limit, then accepted the original mined block. Sustained coordinated skew has not been tested on a valid chain. Actual mining, hostile timestamps and production difficulty remain unverified; no node-uptime rewards are planned.

04 / TRANSACTIONS AND PRIVACY

Privacy remains a release gate.

The live lab exposes sender keys, recipients, amounts and balances. Signed transfers and a limited three-wallet trial confirmed propagation and accounting; a confirmed replay and a signature-tampered transfer were rejected in a bounded live check. These results do not establish anonymity, cross-genesis replay resistance or deep-reorganization safety. The target is private-by-default transactions with publicly verifiable conservation, backed by independent cryptographic review.

05 / SECURITY AND RELEASE

Prove claims before launch.

Native Windows, Linux and macOS packages, safe wallet recovery, adversarial peer tests, durable storage, reproducible builds, an independent audit and an approved fresh genesis remain open. The development chain will be discarded for a production launch. No current test coin, balance or address implies a future mainnet allocation. See the live Security and Transparency report, roadmap and explorer for current evidence.