Frequently Asked Questions

A way to sell USDC for cash where the escrow is released by a cryptographic proof of the buyer's bank payment rather than by a person deciding the payment happened. There is no custodian, no arbitrator, and no dispute path — because there is nothing left to dispute once the proof either exists or does not.

Keys, not logins. There is nothing to sign up for and no password to lose. The protocol does have accounts in the technical sense — an authority policy that decides what may spend, which can be a single key, a threshold of several, or guardian-based recovery — but none of that is a login, and none of it is held by us.

USDC, and only USDC at launch. The deposit path accepts a signed transfer authorisation for one asset, and the approved fee-asset set is USDC from genesis. Adding another asset is a new deployment with its own address rather than a setting we change, so ETH and USDT are roadmap items and are marked as such — not a switch waiting to be flipped.

Settlement is designed for PLN, EUR and USD, and support is per-bank rather than universal: each bank needs its own proof template, because the proof reads specific fields out of that bank's own session. The design targets mBank, Nest Bank and Revolut first, and the list grows as templates land rather than all at once.

The seller moves USDC into an escrow owned by the offer itself — not a platform wallet and not a shared account. From there the offer can do exactly three things: settle against a valid proof, be cancelled by its creator while nobody has taken it, or expire and return the funds. There is no fourth path and no way to add one.

The buyer's browser runs the bank session jointly with an independent notary, which co-signs the encrypted transcript without being able to read it. A zero-knowledge proof then establishes that the transcript contains the right reference and an amount inside the agreed band, and discloses nothing else — not the balance, not the credentials, not the statement.

The protocol that produces that co-signed transcript. It lets one party prove what a secure web session contained without handing over the session itself, using selective disclosure so only the fields that matter are revealed.

A screenshot is a picture of a claim; anyone can edit one, and checking it means a person forming a judgement. A proof is checked by the contract that holds the money, so verification and release are the same event and there is no window where a payment is sent but unconfirmed.

Fees are set by an open market. You bid a fee when you post a trade, a published reserve is the floor, and the auction is first-price — so the bid you make is the fee you pay. The amount and the asset are both public once the trade settles. There is no rate card, and no spread is folded into the exchange rate.

Nobody should quote you a wall-clock number for this. The proving step is bound by bandwidth rather than computation, so a fast connection makes it seconds and a slow one considerably worse; settlement then batches into an epoch and inherits an Ethereum confirmation. The claim is not speed — it is that no human decides.

The amount check is two-sided: the payment must land at or above the agreed figure and no more than a fixed tolerance above it. Overpay past that tolerance and no proof exists at all — the money has left your bank and the escrow expires back to the seller. Inside the band the trade settles, and anything above the agreed amount is gone, because the escrow releases the quantity the offer fixed and there is no operator to run a refund.

Nothing is released. The escrow stays exactly where it is until a proof arrives or the window closes, and if the window closes the funds return to the seller automatically. There is no state in between and no queue to wait in.

Yes, and this is a real cost we do not price away. A taker can claim an offer, never pay, and let it expire — it costs them nothing and it costs the maker the use of that capital until the window closes. There are no claim bonds, so the defence is priced into the spread rather than enforced by the protocol.

No, and not as a policy — as a property. The settlement contract has three functions and no administrator, pause switch or upgrade key, so there is no mechanism by which we or anyone else could move your funds. That forecloses the largest category of loss in the history of bridges and vaults, and it means with equal force that a mistake cannot be undone.

That bank stops working until the notary scheme is replaced. The attestation depends on the bank serving TLS 1.2, TLS Notary has no TLS 1.3 plans, and the response would be a new deployment rather than a migration. It is the single most likely thing to force a redeployment, and it is not under our control.

No — and this is the part most easily oversold. The deposit onto Ethereum is public like any other transaction. What the design provides is unlinkability between that deposit and any later spend. Privacy begins after the deposit, not at it, and anyone claiming otherwise is describing a different system.

Today it is a single, self-hosted notary, and settlement soundness rests on its signature being unforgeable. The known residual risk is collusion between the notary and a prover, which one notary cannot mitigate; the path away from it is a threshold of independent notaries, and that is not built yet.

Access is gone, and nobody gains it — a liveness failure rather than a safety one. Authority is knowledge of a key, and spending also needs the witness data proving what you own. Guardian-based recovery exists precisely for this and has to be configured before it is needed, not after.

Trading crypto peer-to-peer is not prohibited in Poland, but this is not legal advice and the regulatory position is not static — EU-level rules for crypto-asset service providers have been phasing in, and whether a given activity falls under them depends on facts we cannot assess for you. Understand your tax obligations, keep accurate records, and take professional advice for anything consequential.

No. The design is published and the software is not finished. Nothing on this site has processed a real settlement, no figure anywhere on it is a measurement, and the cryptographic work at the centre of it has no existing implementation to borrow from. Treat every claim here as a specification rather than a description.