ERC-8290 Shielded Note Teleport Standard (Shielded Note Teleportation)
Draft · Vaults & DeFi · created 2026-06-05
Defines a way to move shielded notes between UTXO-based privacy protocols to a different pool using only a proof, without publicly withdrawing and redepositing the asset.
Why it is needed
UTXO-based privacy protocols each form an isolated privacy set, so moving an asset from one pool to another requires a public withdrawal followed by a public redeposit. This process leaves a trail that links the asset, amount, source pool, destination pool, timing, and recipient, breaking privacy. Without this standard, users have no choice but to create this kind of public linkage every time they move funds between protocols, and different privacy protocols cannot interoperate unless they share the same note commitment format.
What to implement
function isValidRoot(bytes32 root) external view returns (bool);: Verifies whether the destination pool or a canonical root registry recognizes and trusts a specific canonical root.function teleport(Joinsplit calldata joinsplit) external { require(canonicalRootRegistry.isValidRoot(joinsplit.root), "invalid canonical root"); _verifyTeleportProof(joinsplit); _spendNullifiers(joinsplit.nullifiers); _enforceBackingPolicy(joinsplit); _insertCommitments(joinsplit.commitments); _executeWithdrawals(joinsplit.withdrawals); }: After confirming the canonical root and verifying the teleport proof, consumes the nullifier, applies the backing policy, and either inserts the destination commitment or executes a withdrawal.
In depth
Burn address construction
The burn address is computed as burnAddress = H(chainId, dstPoolAddress, receiver, burnSecret, "ZKTELEPORT"). chainId is the EIP-155 chain identifier, dstPoolAddress is the destination pool address, receiver is the destination owner authorized to receive, and burnSecret is secret entropy chosen for this teleport. Each input is interpreted as a Field inside the circuit, and when exposed as a public input it must be represented as a 32-byte ABI type such as bytes32; if shorter than 32 bytes it must be left-padded with zeros, and if longer it must be reduced to a single field element via canonical encoding or hashing. The burn address must be bound to the destination chain ID, destination pool, and recipient, and cannot be reused for a different chain, pool, or recipient. burnSecret is recommended to include at least 128 bits of private entropy.
Standard source note commitment and export
The standard source pool note commitment can be expressed as blindedOwner = H(ownerAddress, blinding), commitment = H(blindedOwner, data). When exporting a note, the commitment is recomputed by replacing ownerAddress with burnAddress. The teleport proof must prove that this burn commitment is included in the source pool root via Merkle membership (assertMerkleMembership). Protocols with a different note format may define an adapter, which must specify how to compute the burn commitment, how to verify source tree membership, and how to bind the source note to the asset context it is being imported into.
Canonical tree and trust policy
The destination pool must verify that the source pool's commitment root exists in the canonical tree and is valid. The canonical tree can be any authenticated tree trusted by the destination pool, whether that is a chain's state tree or a custom Merkle tree. For a custom Merkle tree, the canonical tree proof is expressed as Merkle membership showing that canonicalLeaf is included in canonicalRoot. The destination pool must reject any teleport proof referencing a canonical root that it, or its configured canonical root registry, does not recognize, and the registry is recommended to expose a lookup method such as isValidRoot(bytes32 root).
Teleport import procedure
On import, the destination pool must verify a teleport proof that includes as public inputs the canonicalRoot, the source context, the asset context, one or more nullifiers, one or more destination commitments or withdrawals, the destination operation digest, and sufficient transaction authorization data to bind the proof to the submitted import operation. Each nullifier must be marked as used before or during import finalization, and the same nullifier must not be accepted more than once by the same destination pool. The destination pool must, per protocol rules, insert the imported commitment into its own commitment tree, execute the destination withdrawal, or both.
Nullifiers and replay protection
A teleport proof must expose one or more nullifiers unique to the teleported note, according to the destination pool's nullification rules, and each nullifier must be derived from private material or a nullifying key associated with the teleported note. The mechanism must guarantee that the same teleport note is not imported into the same destination pool more than once. It is recommended that destination pools domain-separate nullifiers per destination pool, or track them only within the destination pool where the proof is consumed; a proof generated for one destination pool must not be replayable on another.
Backing model and adapter responsibilities
The destination pool must define a backing model for imported value before accepting a teleport proof. This model may use escrowed liquidity, burn-and-mint accounting, protocol-owned inventory, a bridge settlement mechanism, or another explicit policy, and it must not create destination notes or execute withdrawals unless backed under that policy. Adapters must define the source chain identifier, the source commitment hash, the source tree membership hasher and depth, the canonical tree proof representation, the asset context encoding, and the source burn/lock conditions, and must not weaken destination binding, recipient binding, source root membership, canonical tree membership, or asset context preservation.
Authorization
It is recommended that destination pools bind each import to explicit user authorization, such as an EIP-712 typed data signature. When typed data authorization is used, the circuit must verify that the signer matches the recipient bound to the burn address, and the destination contract must verify that the proof exposes the same typed data digest that the contract accepts.
Proposal status
This proposal is currently in Draft status, and the detailed specification may still change or be abandoned.
Things to watch out for
- Blast radius of proof system failures: Teleport safety depends on the robustness of the source membership proof, the canonical tree proof, and the destination import proof; a failure in any one of these can lead to inflation, theft, or replay.
- Reusability of burned notes: The destination pool must confirm that the imported value is backed by a note that is locked or burned under the source protocol's rules and can no longer be spent; if the source protocol cannot prevent further spending of a burned note, the destination protocol must handle that risk directly.
- Protecting the burn secret: The burn secret must remain private until the teleport proof is generated, and reusing the same burn secret across multiple teleports is discouraged.
- Nullifiers and cross-replay: The destination pool must prevent replay through nullifier tracking, and must also bind the burn address and authorization message to the destination pool to prevent cross-replay into a different destination pool.
- Adapter implementation errors: If an adapter fails to preserve the exact commitment semantics of the source protocol, errors in hasher selection, field truncation, byte ordering, token encoding, or tree depth can produce weak proofs or make it impossible to import a valid note.