Zakura, the Zcash full-node software team, published a proposal titled Quantum Unlinkability for Zcash Swaps on X on Oct. 18 Taipei time. Written by engineer Adam Tucker, the paper says ZEC cross-chain swaps can gain quantum-resistant unlinkability without a network upgrade. Its core warning is aimed at NEAR Intents: refund and receiving shielded addresses used in swap flows are written to the public NEAR blockchain, creating a record that could become sensitive if quantum computers later gain the ability to reverse the cryptographic assumptions behind those addresses.
BlockTempo said it could not find any formal cooperation agreement, strategic binding arrangement, or funding relationship between ZEC and NEAR. Even so, the report said wallet-based swap activity already relies heavily on NEAR Intents. Zakura itself is being developed through a collaboration between Project Tachyon, led by Zcash co-founder Sean Bowe, and Valar Group, led by Osmosis co-founder Dev Ojha. The team said its funding comes from private ZEC donations.
Why a public refund address matters
According to Zakura, all addresses in a Zcash shielded account are derived from the same incoming viewing key, or ivk, which is used to identify and decrypt incoming payments. The addresses may look different, but they are generated from that key through elliptic-curve operations. On classical computers, deriving the key from an address is blocked by the discrete logarithm problem. A quantum computer capable of running Shor’s algorithm could change that assumption.
If that happened, an attacker could see every historical incoming payment to the account, including amounts and memos, and link them together. The attacker still would not be able to spend the funds, because spending requires a different key.
Sean Bowe previously wrote on his personal blog that Zcash shielded transactions already have quantum-resistant privacy in many common situations, as long as an attacker does not have your payment address. He also said that if elliptic curves are broken, the threat goes beyond privacy and reaches protocol soundness, including the possibility of forged or stolen funds. He said the team plans to remove “in-band secret distribution” in Tachyon to defend against quantum privacy attacks more broadly. In that framing, address exposure remains the weak point, which is exactly what a “harvest now, decrypt later” strategy would wait for.
NEAR Intents is identified as the source of the exposure
Zakura said the issue comes from the way NEAR Intents handles swap flows. When a user swaps ZEC for assets on another chain, the quote includes a shielded refundTo address. Once the deposit is registered, that address is written to the public NEAR chain whether or not a refund is ever sent. In the reverse direction, when assets are swapped into ZEC, the receiving address is also exposed. Zakura’s engineering page describes these addresses as being published hundreds of times per day.
The concern is not limited to a single refund event. The problem is that these addresses accumulate on a public chain over time. If quantum attacks become practical in the future, those records could be used to reconstruct a wallet’s incoming payment history.
One-time ivk and PIR form the proposed fix
Zakura’s answer is a “one-time ivk” design. For each swap, the wallet re-derives only the randomized parameter rivk inside the account key material, producing a separate ivk and a separate address for that swap, while the spending-related keys stay unchanged. The funds remain in the same account.
Under that design, Zakura said a quantum attacker who breaks one exposed address would only learn about that single refund. To trace the address back to the main account, the attacker would also need to reverse the hash derivation step, which Zakura argues is also out of reach for quantum computers.
The team said this avoids the drawbacks of opening a new account for every swap. Zakura pointed to hardware constraints as part of the reason: Keystone can sign only account 0, and Ledger can use only a single account per transaction.
The tradeoff is wallet recovery. Restoring from a seed phrase would require scanning many more keys. Zakura described a two-layer approach to handle that:
- The wallet leaves a memo in change outputs that only it can read, recording the index used for each swap.
- Because the NEAR bridge encrypts its Zcash-side payouts with an all-zero outgoing viewing key, anyone can build an address-indexed database of NEAR payouts, and wallets can query that database through Private Information Retrieval, or PIR, so the server does not know what the wallet searched for.
Zakura said wallets that adopt its open-source library, including Vizor, could integrate the design in the coming weeks. Hardware wallets would need firmware support first. The engineering page currently marks the item as “under review.”
No formal partnership found, but product integration is extensive
BlockTempo said it found no formal strategic cooperation or binding agreement between Zcash development organizations and NEAR. The connection, it said, comes mainly from product integration.
Electric Coin Company rolled out CrossPay and swap deposits in the Zashi wallet in 2025, and both features are powered by NEAR Intents. After the original team reorganized into ZODL this year, the official claim was that more than $600 million in ZEC swaps had been facilitated since October 2025. Vizor, built by the Keplr team, also labels its swap feature as powered by NEAR Intents.
On the NEAR side, an official blog post in July said more than $1.5 billion in ZEC had already been settled through NEAR Intents. According to NEAR Intents Explorer, as of 8:58 p.m. Taipei time on Oct. 19, cumulative ZEC volume was about $2.54 billion and 30-day volume was about $263 million, accounting for about 7.9% and 9.0% of the public total, respectively, based on BlockTempo’s calculation.
The report also said the NEAR Intents team had previously applied for retroactive funding on the Zcash forum and said it wanted deeper cooperation, then later withdrew the application while noting that ZEC had been bridged to Solana. CoinGecko data cited in the report showed ZEC at about $1,140 and NEAR at about $5.00 at the same time.
Privacy, custody, and permission debates remain unresolved
That dependency has also drawn criticism. Zcash community member emersonian warned in February on the same forum thread about “traceable Orchard,” arguing that deposits and withdrawals can still be tracked when the amounts and timing are close. He also said NEAR validators custody ZEC through TSS, and that real shielded support may require FROST instead.
Other community members questioned the use of TRM Labs and other AML data sources in 1Click API quote screening, saying the privacy impact had not been explained.
BlockTempo also cited CoinDesk reporting that NEAR Intents used its SHIELD system in September to block funds tied to the Bitget hack, which triggered debate over whether the system is truly permissionless. On Oct. 1, the Omni deposit and withdrawal vulnerability led to the theft of about $3.8 million. The funds were later fully recovered, and the official investigation was halted.
The Zakura proposal does more than outline a technical patch. It also puts a spotlight on how much ZEC swap activity now runs through NEAR Intents, and on the privacy and trust costs that come with that dependence.


