Draft Design of RGB Minimal Compliance Asset Verification Mechanism


Introduction: Balancing Privacy and Compliance
In the process of tokenizing financial assets on-chain, achieving regulatory-verifiable compliance often conflicts with user privacy protection. The RGB protocol, as a Layer 2 smart contract system on Bitcoin, employs a client-side validation model combined with on-chain commitments to enable highly private asset transfers: by committing hash values of off-chain transaction data to the Bitcoin main chain, RGB clients can verify historical data provided by others, while third parties only see hash commitments and cannot reconstruct the original content, thereby safeguarding privacy. This means that, by default, external observers (including regulators) cannot directly access asset flow paths or details. However, compliance scenarios require that authorized parties be able to audit asset transaction chains. Therefore, we need to design a "minimal compliance" mechanism: enabling regulators or authorized third parties (viewing parties) to verify the transaction paths and compliance of specific assets while minimizing sacrifices to privacy.
This proposal outlines two implementation paths:
the first relies entirely on RGB contract Schema extensions to achieve compliance control,
the second embeds compliance information in the L1 layer (BTC/Tondi chain) transaction structure and combines it with L2 client validation.
Both approaches emphasize the principle of minimal changes and ensure that malicious clients cannot evade compliance constraints (i.e., they possess anti-client cheating capabilities).
Scheme One: Off-Chain Compliance Mechanism Based on RGB Schema
Scheme Overview
In this scheme, compliance requirements are implemented by extending the RGB contract's Schema and validation scripts, without modifying the Bitcoin main chain protocol. The core idea is to add compliance-related global states and transfer validation rules to the RGB smart contract, ensuring that each asset transfer includes regulatory permission proofs and whitelist checks. During local validation of asset transfers, clients will inspect these compliance fields, rendering any transfers without proper regulatory signatures or to blacklisted recipients invalid.
graph TD
A[User Initiates Asset Transfer] --> B[Generate Regulatory Viewing Field - Encrypt Transaction Details with Viewing Key]
B --> C[Obtain Regulatory Signature - Send Transfer Info to Regulator for Approval]
C --> D[Include Whitelist Proof - Provide Merkle Path for Recipient]
D --> E[Compute State Commitment - Hash All Data Including Compliance Fields]
E --> F[Anchor to L1 Chain - Commit Hash via OP_RETURN or Taproot]
F --> G[Recipient Validates - Check Signature, Whitelist, and Commitment Consistency]
G -->|Valid| H[Accept Transfer]
G -->|Invalid| I[Reject Transfer]
Viewing Key Integration and Viewing Mechanism
Under RGB's off-chain model, a "viewing key" concept similar to zero-knowledge chains (e.g., Zcash) can be introduced. Specifically, the regulator/viewing party's public key is predefined in the contract, and this key is used to encrypt key information for each transfer, generating a summary that can only be decrypted with the corresponding private key. For instance, a "regulatory viewing field" is added to the RGB transfer state, containing transaction details (such as hashes of sender and receiver identifiers, and asset amounts) encrypted with the regulatory public key. This encrypted field is hashed into the on-chain commitment as metadata for the transfer operation, ensuring immutability and strong binding to the transfer.
In practice, when a user initiates an asset transfer, the client will: ① Generate a message with transaction details and encrypt it using the regulatory viewing public key; ② Include the ciphertext in the RGB transfer operation data package and compute the overall state commitment. The on-chain anchoring transaction publicly records this commitment hash (see below). For regulatory viewing, the regulator holds the corresponding viewing private key and can extract and decrypt the plaintext from the off-chain transaction data provided by the user (or pre-shared materials) to verify if the transaction details meet requirements. If the regulator operates its own RGB node, it can require real-time access to these transfer data packages and decrypt them independently for auditing. Since the viewing key only grants read permissions and not spending permissions, it enables "selective disclosure" of transaction information without endangering user asset control, thus satisfying audit needs while preserving on-chain privacy. Note that under the RGB model, regulators typically cannot passively obtain complete plaintext data from the chain and still require asset issuers or holders to provide off-chain data packages for interpretation. This can be ensured through institutional norms or business processes (e.g., requiring users to file encrypted transaction records with regulators after transfers).
Strong Binding Mechanism for Regulatory Signature Authorization
To ensure each asset transfer obtains prior approval, this scheme introduces a regulatory signature field in the RGB contract transfer operation and enforces its validity via contract validation scripts. Specifically, in the RGB Schema's transfer operation (Transition), add a metadata field such as reg_signature, typed as an ECDSA/Schnorr signature. The contract's validation script mandates that reg_signature must be a valid signature from the regulator's private key (with the public key pre-stored in the contract's global state) over the transfer summary. The transfer summary may include key transaction information (e.g., asset ID, sender and receiver Bitcoin UTXO identifiers, transfer amount, previous state hash), ensuring each transfer request has a unique corresponding signature message. The strong binding mechanism is evident: the signature is incorporated into the on-chain commitment calculation and subject to RGB script validation; if missing or invalid, any client adhering to the contract will reject the state transition. This is equivalent to requiring regulatory "signature and seal" before asset ownership can be transferred. Since the signature is based on standard cryptographic algorithms, clients cannot forge regulatory authorization, preventing users from bypassing approval for private transfers (anti-cheating). This design parallels Blockstream's AMP platform on the Liquid network: AMP enforces whitelist control by requiring the issuer's server to co-sign each transfer, only doing so after confirming the recipient's compliance. Our scheme embeds this "regulatory co-signing" logic into the RGB contract via digital signature validation, making regulatory authorization technically immutable and unavoidable.
In implementation details, the regulator's public key RegPK must be preset in the RGB contract's global state (definable at contract genesis). During a transfer, the user first sends the proposed transfer information to the regulator for approval, who signs the summary with their private key to generate s = Sign(RegSK, msg). The user fills s into the transfer operation metadata and broadcasts it to the recipient. The recipient's client verifies the signature using the known RegPK and checks that the message matches the actual transfer content. Once the signature and other state validations pass, the recipient can accept the asset. This mechanism ensures that without a regulatory signature, the transaction is invalid, achieving one-to-one strong binding between regulatory authorization and asset transfer.
Whitelist Public Key/Hash Binding
To restrict asset circulation only among permitted addresses/entities, we introduce a whitelist concept. Implementation options include static and dynamic whitelists. Under the static scheme, store a whitelist hash (e.g., Merkle tree root) in the RGB contract's global state. This whitelist is set by the issuer/regulator at asset issuance, containing hashes of all permitted holders' user identifiers (e.g., Bitcoin public key hashes, RGB user IDs). Subsequently, in each transfer's validation script, require proof of the new recipient's inclusion: e.g., add a whitelist_proof field in the transfer metadata with the recipient's Merkle path. The validation script loads the global whitelist root, computes verification with the provided leaf hash and path, confirming the recipient is on the list. If verification fails, the transfer is invalid. Thus, even if sender and recipient collude to bypass regulation by sending to unverified/unpermitted users, the recipient's client will detect the whitelist failure and reject the asset. Under the dynamic scheme, allow regulators to update the whitelist via special transactions (e.g., add/remove users); the whitelist root evolves as a global state, updated by regulator-signed change operations, with each update committed on-chain and notified to all users.
Whitelist constraints ensure asset flows are limited to participants meeting specific conditions, akin to restricted securities or qualified investor regimes in securities markets. For example, Blockstream AMP limits Liquid tokens to recipients registered in the AMP system and meeting category requirements. Our RGB scheme implements this restriction directly at the contract layer via cryptographic proofs, enabling any client to independently verify compliance without centralized queries.
Embedding Compliance Identifiers in Anchoring Transactions
Under a pure RGB scheme, on-chain anchoring transactions primarily carry RGB state commitments, with the Bitcoin main chain unaware of the commitment content. To facilitate quick identification by regulators or observer nodes that "this Bitcoin transaction involves a compliant asset transfer," we can embed special identifiers in the anchoring transaction. Possible methods include:
OP_RETURN Prefix Identifier: If RGB uses OP_RETURN to write commitments on-chain (RGB supports saving commitments via OP_RETURN), add a specific prefix to the OP_RETURN data. For example, prefix with bytes indicating "compliant asset transfer" (e.g., 0x434F4D50 for "COMP"). Ordinary nodes seeing this magic number at the OP_RETURN start can judge it as a compliant RGB anchoring transaction. Note OP_RETURN byte limits and that prefixes may reduce commitment data entropy, but with proper design (e.g., hashing prefix concatenated with commitment), security is maintained.
Taproot Script Leaf Identifier: RGB also supports embedding commitments in Taproot output script trees (i.e., "Tapret" method). Add a fixed-format leaf node script as a compliance flag in the Taproot Merkle tree, e.g., a leaf script as OP_RETURN (containing regulator identifier or simple constant). This script is never spent (only as commitment), but its presence affects the Taproot output public key hash. When generating the Taproot address, the internal public key is tweaked with the entire script tree hash. Since the leaf is fixed and known, its existence can be verified offline: regulators or observers, knowing the output's original public key and fixed leaf script hash, can compute the expected tweaked public key and compare it to the on-chain output to confirm inclusion. During regular spending, users use the Taproot key path (without revealing scripts), so the identifier remains hidden to the public but verifiable by regulators with extra information. This equates to subtly attaching a "compliance watermark" to each anchoring output.
Transaction Version or Type Field: Standard Bitcoin transaction formats lack explicit "type" fields, but on new chains (e.g., Tondi), extended formats could reserve bits for marking transaction types, such as an unused nVersion bit for "compliant asset transaction." Ordinary nodes could quickly classify based on this. However, this is inapplicable to BTC mainnet and serves only as a custom idea for new chains.
In Scheme One, since we make no changes to the Bitcoin protocol, the above identifier methods use existing script mechanisms (OP_RETURN or Taproot trees). The Taproot scheme is particularly suitable for hidden marking: it leverages Taproot's tweak mechanism to embed flag data with zero extra bytes, aligning with the "minimal change" spirit. RGB already uses similar principles in Tapret, placing commitment data in Taproot script leaves before hashing into public keys. Building on this, adding a fixed flag leaf simultaneously identifies the transaction as regulated. Compliance identifiers in anchoring transactions enable regulators to scan the blockchain for RGB transactions and pinpoint those involving regulated assets (assuming regulators know the identifier content or possess scanning keys).
Anti-Client Cheating and Security
Scheme One eliminates typical cheating avenues through cryptographic validation and protocol rules, ensuring malicious users cannot bypass compliance constraints by modifying clients:
No Transfer Without Signature: Regulatory signatures are mandatorily validated; without one, the transaction is invalid at the contract layer and cannot deceive recipients. Signatures are hard to forge, ensuring only authorized transactions succeed.
No Receipt for Non-Whitelisted: Whitelist validation ensures non-compliant recipient transactions are also invalid. Senders cannot conceal recipient identities, as recipients must provide proof of listing to acknowledge assets.
Commitment Binding Against Tampering: All compliance data (signatures, whitelist proofs, viewing ciphertexts) are included in on-chain commitments. Attempting to remove or alter these fields changes the commitment hash, which recipients will detect as inconsistent and reject. Just as RGB ensures client validation of off-chain data against on-chain commitments to prevent fakes, compliance fields are part of this consistency check.
On-Chain Marking Aids Regulation: While markings do not confer transaction validity checks (they are auxiliary), they enhance auditability. Regulators can scan for markings to detect suspicious transactions without required markings (e.g., those that should have them but don't, possibly indicating covert transfers). Combined with business rules (e.g., mandating marked formats for all compliant asset transfers), irregular transactions can be identified.
In summary, Scheme One implements compliance control solely at the RGB contract layer, transparent to ordinary users and miners without L1 consensus changes. It ensures assets can only transfer successfully between clients if compliance field requirements are met, technically enforcing regulatory provisions. Its drawback is that regulatory audits still rely on obtaining off-chain data (users must submit materials or regulators participate in each signature process for copies). To address this, Scheme Two explores L1-assisted data acquisition and validation methods.
Scheme Two: Collaborative Compliance Mechanism with L1 Transaction Structure + L2 Validation
Scheme Overview
Scheme Two combines modifications to the underlying blockchain transaction structure with RGB client validation to implement compliance control at both L1 and L2 layers. The idea is to directly embed or include compliance-required information (e.g., regulatory signatures or identifiers) in Bitcoin or Tondi main chain transaction formats, making it easier for regulators to access and verify. Meanwhile, RGB L2 contracts still handle detailed validation and anti-cheating. Through L1-L2 synergy, this mechanism aims to reduce reliance on off-chain data, allowing regulators to extract partial verification evidence directly from the blockchain. Design adheres to minimal change principles, leveraging existing L1 features (e.g., Taproot, scripts) to accommodate new information.
graph TD
A[User Initiates Asset Transfer] --> B[Generate Encrypted Details - Viewing Key Encryption for L1 Embedding]
B --> C[Obtain Regulatory Signature - Include in L1 Transaction Structure]
C --> D[Embed Whitelist Proof/Hash - Via OP_RETURN, Witness, or Taproot]
D --> E[Construct Anchoring Transaction - With Compliance Identifiers in L1]
E --> F[Commit to Chain - L1]
F --> G[Recipient Validates in L2 - Cross-Check L1 Data with RGB Rules]
G -->|Valid| H[Accept Transfer]
G -->|Invalid| I[Reject Transfer]
J[Regulator Audits] --> K[Extract from L1 - Decrypt Ciphertexts, Verify Signatures]
Modifications and Extensions to L1 Transaction Structure
Directly modifying transaction structures on Bitcoin mainnet (e.g., adding fields) requires a soft fork, which is costly and unrealistic short-term. However, we can leverage existing flexible fields to embed compliance data:
Adding Auxiliary Outputs: In anchoring transactions, add a special output for recording compliance data, e.g., a zero-value OP_RETURN output storing regulator-related information (encrypted or hashed). Conventional RGB anchoring may already use one OP_RETURN for commitments; if multiple are allowed, dedicate a second to compliance markings and data. Thus, regulators can read this output directly from the blockchain to extract needed info (e.g., signature values or encrypted payloads) without off-chain data. This does not affect UTXOs (OP_RETURN is unspendable) and is backward-compatible. Note Bitcoin's default 80-byte OP_RETURN limit, requiring careful data organization, such as compressed or truncated regulatory signatures or hash references.
Extending Witness Data: SegWit transactions have witness areas for arbitrary data, transparent to non-witness systems. Consider placing compliance proof data in an input's scriptWitness, e.g., having one input's witness carry the regulatory signature and whitelist proof as its first item after a fixed tag. For verification, regulators check if the transaction input contains expected-format data. This makes data propagate with the transaction without extra outputs. However, parsing is slightly complex, needing consensus or standard rules for witness formats; otherwise, full nodes won't parse specific witness content. Thus, this suits regulator-owned nodes or custom parsing tools.
Taproot Tweaks: Extending Scheme One's Taproot script tree idea, fully leverage Taproot flexibility. Specifically, embed regulatory signatures or authorization proofs into the Taproot output public key itself. Bitcoin Taproot output public keys are computed as P_out = P_internal + H(P_internal || merkle_root)*G. If regulatory data (e.g., signature hash or authorization code) is part of the Merkle tree or directly in hash computation, the output public key effectively "engraves" this data. This requires a specific Taproot internal structure: e.g., one path for normal RGB commitment leaves, another for regulatory data leaves (or multi-leaf combined Merkle roots). The final output public key commits to both. For verification, regulators require users to provide Taproot internals (internal public key and Merkle proofs, akin to RGB's ETP extra transaction proofs) to independently compute and validate included regulatory data. Advantages: On-chain, it's just a Taproot output with no extra outputs or OP_RETURN; data is hidden from the public but verifiable by informed auditors. Lightning Labs' Taproot Assets protocol uses similar "taptweak" techniques, twisting asset data hashes into Taproot keys for zero extra block burden.
On new L1s like Tondi, we could directly introduce new transaction structure elements to simplify. For example, Tondi or innovative chains could define dedicated "compliance transaction types" with extra Metadata fields for issuers/regulators to write notes; or enforce certain asset transactions to include regulatory signatures at the consensus layer. Given GhostDAG's potential for higher throughput and concurrency, accommodating minor extra data is feasible. Without deep Tondi specifics, we assume it at least matches Bitcoin's script and Taproot capabilities, allowing embeddings as above.
On-Chain Embedding of Viewing Key and Plaintext Data
Scheme Two aims for more autonomous regulatory viewing, so consider storing certain encrypted transaction plaintexts directly on L1. Specifically, place transaction details encrypted with the regulatory viewing key in anchoring transaction auxiliary outputs or Taproot scripts. For example: Encrypt the transfer's plaintext (including sender/receiver identity identifiers and amounts) with the regulator's public key to get ciphertext C, then write C to the transaction's OP_RETURN or as a Taproot leaf content. This ciphertext becomes public blockchain data, but since it's encrypted with the regulatory public key, only the regulatory private key holder can decrypt to plaintext. Others see only random bytes, preserving privacy.
For instance, on Bitcoin mainnet, construct an OP_RETURN output as: OP_RETURN , where identifies it as compliance viewing data, and is the ciphertext. For each compliant asset transfer, this OP_RETURN appears, allowing regulatory nodes to monitor the feature, extract ciphertext, and decrypt with their private key for transaction details. Since the ciphertext is part of the RGB state commitment, it ensures consistency with actual off-chain data (consistency provided by RGB proofs), preventing users from uploading mismatched fake ciphertexts. This mirrors some privacy coins' on-chain recording of controlled plaintexts: e.g., Zcash stores encrypted memos and amounts on-chain, decodable only with viewing keys. Our scheme publishes key transaction info ciphertexts alongside anchoring transactions, enabling regulators to independently obtain audit plaintext without requesting user data.
Note Bitcoin's limited OP_RETURN capacity may constrain encryptable info size; opt to encrypt only essential identifiers rather than full data, e.g., previous transfer hash, current recipient ID, and asset amount for chaining asset flow paths. Regulators can iteratively decrypt each transfer's ciphertext to reconstruct paths from issuance to present (like on-chain tracking, but info visible only to regulators). For anti-abuse, since ciphertexts are unreadable except by regulators, the public gains no sensitive info, and OP_RETURN data doesn't affect UTXO sets, minimizing network impact.
In Taproot scenarios, embed ciphertexts in unspent script branches, eliminating even OP_RETURN outputs but requiring regulators to have users provide script branch reveal proofs for access. For self-identification and acquisition by regulatory nodes, OP_RETURN is more direct.
Integration of Regulatory Signatures in L1 & L2
In Scheme Two, regulatory signatures can be used for L2 validation and potentially participate directly in L1. For example, multi-sig ideas: Set anchoring UTXOs under joint user+regulator control. But this alters transfer flows and introduces custody risks, not "minimal change." Instead, opt for on-chain endorsement of regulatory signatures without actual joint control.
A compromise: Regulatory signatures as on-chain endorsements. The regulator issues a signature for each transfer and includes it directly in the anchoring transaction (via OP_RETURN or witness data). Miners and nodes need not validate it (unless consensus rules, which we don't assume), but regulators or observers can read and verify it on-chain. For example, anchoring transactions include OP_RETURN ; all see the signature string, but only those knowing its meaning can validate (requiring knowledge of the signed message—perhaps the transaction's identifier). RGB clients still validate signature correctness in L2, but now data comes from on-chain, not sender-provided. This enhances reliability—even if senders maliciously withhold signatures, recipients can fetch them from the blockchain and verify.
Publishing signatures plaintext on-chain has privacy costs: While signatures don't directly reveal transfer details, they may leak patterns (fixed lengths are negligible). If the signature message is public info like txid, the signature reveals no recipient identity, minimizing risk. If including private info, sign only hashes. Overall, treating regulatory signature results as first-class on-chain data allows direct perception of regulatory authorization by chain participants. It resembles traditional finance's regulator-stamped vouchers filed publicly with transactions, but digitally.
On-Chain Assistance for Whitelist Control
While L2 contracts handle whitelist validation, L1 can assist. For example, on-chain whitelist hashes: When regulators publish or update whitelists, send special on-chain transactions with new Merkle roots, effective heights, etc. (via OP_RETURN announcements). Anyone (especially clients) can fetch latest roots from the chain, ensuring sync with regulators without offline notifications. For each asset anchoring transaction, clients validate whitelist proofs in L2 and cross-check current roots against on-chain announcements, preventing malicious users from forging expired or incorrect proofs.
Another direction: L1 receiver address whitelisting, e.g., scripts restricting recipient UTXOs to specific prefixes or public keys. Bitcoin scripts don't support direct "address in list" checks, but new chains (Tondi) could extend script capabilities (e.g., Merkle check opcodes). If Tondi supports more complex Scripts, outputs could require regulatory signatures and whitelist Merkle proofs for unlocking, front-loading compliance checks on-chain. However, this increases complexity and costs, so prefer L2 validation with on-chain auxiliary records.
Compliance Identifiers in Anchoring Transactions (Taproot Tweak / Transaction Type / Script Leaf)
Scheme Two emphasizes anchoring transaction identifiability. Compared to Scheme One, we aim for transactions to carry more compliance info, enriching identifier methods:
For Bitcoin Mainnet: Combine Taproot tweak and script leaf methods. For example, design Taproot outputs with two Merkle leaves: one for RGB commitments, another for predefined compliance flags and possible ciphertexts. This dual-leaf structure tweaks the output public key to cover both. Holders of internal keys and leaf info can verify both commitments. To outsiders, it appears as ordinary Taproot (preserving privacy). Since Bitcoin lacks transaction type fields, simulate equivalents via script patterns (e.g., OP_RETURN outputs, special Taproot leaves).
For Tondi Chain: If Tondi provides transaction types or extra fields, directly record compliance flags as transaction metadata, visible to all nodes. This is stronger than Bitcoin. For example, Tondi or GhostDAG chains might support varied formats, marking "asset transfer transaction" or attaching "RegAuthSig: " in headers. Block explorers or node RPCs could identify and extract compliance info directly. Even without, Tondi at least matches Bitcoin's script/Taproot, allowing similar embeddings. Given GhostDAG's concurrent blocks, Tondi may handle marked transactions better performance-wise, but this doesn't affect embedding methods.
Commonalities: BTC and Tondi are UTXO models supporting Taproot, so RGB asset commitment anchoring is analogous. RGB's anchor abstraction decouples from underlying chains, requiring only basic "embed data in transaction outputs." Bitcoin successfully uses OP_RETURN and Taproot branches for commitments; Tondi likely does too. Thus, for both schemes, most compliance components (viewing key encryption, regulatory signatures, whitelist proofs) can record via OP_RETURN or Taproot commitments in BTC/Tondi anchoring transactions, without restructuring RGB logic.
Bitcoin's Limitations: Bitcoin mainnet is stringent on non-payment data, with OP_RETURN size limits (~80 bytes) and Script complexity caps (OP_CODES execution not unlimited) constraining embeddable info. Lacking built-in asset concepts or transaction classifications, all compliance identifiers must be simulated (e.g., via specific script patterns). Implementing Scheme Two on Bitcoin requires methods (OP_RETURN count, Taproot leaf quantity) to comply with consensus and standards, or risk miner rejection/unnecessary alerts. E.g., multiple OP_RETURNs are consensus-allowed but non-standard transactions may be refused, needing special handling. Taproot leaf use is fully hidden, non-disruptive to networks, but verifying leaves requires extra ETP proofs.
Tondi's Opportunities: As a next-gen L1, Tondi ("copper coin" chain) reportedly uses GhostDAG consensus for improved concurrency and confirmation speed. Its name emphasizes Taproot, supporting Bitcoin's Taproot/Schnorr features. This means our compliance mechanisms are feasible on Tondi and potentially optimizable via protocol upgrades. E.g., Tondi could add type fields for marking asset transactions; allow larger OP_RETURN (with bigger blocks); or provide new Script ops for direct signature validation, shifting some compliance from clients to on-chain (e.g., OP_CHECKSIGFROMSTACK-like for regulatory signatures). If realized, Tondi RGB compliance schemes would be simpler. Even without special support, adopt Bitcoin-equivalent methods. GhostDAG concurrency may introduce transaction order uncertainty, but RGB states depend on specific UTXO sequences, unaffected by block DAG structures—as long as final confirmation order is consistent, RGB validation proceeds normally. Thus, GhostDAG mainly impacts confirmation speed, benefiting real-time regulatory monitoring. In summary, BTC and Tondi support our anchoring and compliance structures similarly; Tondi has potential for protocol evolution to make mechanisms more efficient and intuitive, but our design doesn't strongly depend on that. Implementation differences mainly concern capacity and standardization. E.g., on Tondi, confidently place regulatory ciphertexts in slightly larger OP_RETURN without packing issues; on BTC, prefer compact Taproot leaves to avoid limits.
Security and Minimal Change Analysis
Scheme Two reduces reliance on off-chain communication and trust by on-chaining partial compliance data, while maintaining minimal invasiveness:
Minimal Changes: We only utilize existing transaction outputs and witness fields, without requiring Bitcoin consensus modifications. All embeddings are within script semantics (e.g., OP_RETURN, Taproot commitments), unaffected for non-compliant users. For new chains like Tondi, limit changes to auxiliary means, not forcing network-wide complex validation, preserving main chain simplicity and efficiency.
On-Chain Evidence and Privacy: On-chaining regulatory signatures, encrypted payloads provides transparent regulatory evidence: Transactions carry "regulator-approved" credentials. This aids future dispute resolution and compliance audits. Since key data is hashed or encrypted, privacy leakage is minimal—public sees no plaintext transaction content, only signatures/ciphertexts exist, protecting user details. This achieves a "privacy to the masses, transparency to regulators" balance.
Anti-Client Cheating: Beyond RGB validation, Scheme Two adds on-chain constraints, making cheating harder. E.g., if users attempt transfers without regulatory signatures, RGB validation rejects, and on-chain lacks signature records, allowing easy anomaly detection (missing OP_RETURN or specific witnesses). This forms dual safeguards. If reusing old signatures on-chain, including transaction-specific info (e.g., unique UTXOs or timestamps) makes new transaction hashes differ, invalidating signatures via RGB and showing mismatches on-chain. Overall, under dual L1-L2 constraints, cheating requires deceiving both layers, nearly impossible without detection.
Performance and Overhead: Scheme Two slightly increases on-chain overhead (extra outputs or data), but with small data volumes (e.g., 64-byte signatures, 32-byte hashes) possibly in witnesses, block capacity impact is controllable. Taproot utilization means even added script leaves spend via key paths, adding no runtime overhead. Thus, it meets "minimal compliance" requirements—adding only necessary compliance info.
In summary, Scheme Two offers stronger on-chain verifiability than Scheme One: Regulators need not fully rely on parties for materials and can extract proofs from the blockchain to check if asset flows follow rules. This is crucial for frequent audits or large-scale regulation. It still retains RGB's privacy advantages, hiding details from unauthorized parties.
BTC vs. Tondi Adaptation Differences and Commonalities
Commonalities in Anchoring Structure Support: Bitcoin and Tondi share UTXO models and Taproot support, enabling analogous RGB asset commitment anchoring. RGB's design decouples commitment schemes from underlying chains, needing only basic data embedding in transaction outputs. Bitcoin uses OP_RETURN and Taproot branches successfully; Tondi should too. Thus, for both schemes, compliance mechanisms are recordable via OP_RETURN or Taproot commitments in BTC/Tondi anchoring transactions, without RGB logic rework.
Bitcoin's Constraints: As above, Bitcoin's non-payment data restrictions and lack of asset/transaction classification force simulated identifiers.
Tondi's Potential: As detailed, Tondi's GhostDAG and Taproot focus offer optimization opportunities, though our design uses equivalent methods if unavailable.
Conclusion: Minimal Changes and Anti-Cheating Capabilities
In conclusion, both proposed mechanisms adhere to minimal compliance principles, implementing compliant controls on asset flows without disrupting existing protocols or privacy advantages:
Scheme One (RGB Schema Extension): Fully implemented at Layer 2, with zero intrusion to Bitcoin or Tondi chain nodes. Compliance logic as contract rules offers high flexibility for regulatory customization. Data is entirely off-chain, bound by hash commitments, ensuring privacy. For users, besides obtaining extra regulatory signatures, usage mirrors ordinary RGB assets, with no added on-chain fees. Minimal changes: Only adjust issued asset Schemas and client software; on-chain stores OP_RETURN or Taproot commitments as usual, without new transaction types. Anti-cheating is guaranteed by contract scripts—as long as clients follow RGB validation, non-compliant transfers cannot deceive recipients. If users transact without RGB validation, ignoring compliance fields, acquired assets are deemed invalid in formal compliant networks, losing value. This deters cheaters: Non-rule adherence isolates from the network.
Scheme Two (L1+L2 Synergy): Retains RGB off-chain validation while on-chaining key info (signatures, ciphertexts) to boost transparency and audit ease. Clever use of existing on-chain extensions (Taproot tweaks, OP_RETURN) makes it "microsurgery"-level changes, without hard forks. Even for new chains like Tondi, propose improvements as optional enhancements, not mandatory consensus alterations. Core idea: Leave "regulatory footprints" on the blockchain for dual on/off-chain validation, raising cheating difficulty and irregularity detection. Even if client tampers off-chain records, on-chain signatures/identifiers expose them; transactions without on-chain endorsements are automatically suspicious. This dual mechanism ensures timely discovery or technical prevention of bypass attempts.
Finally, both schemes respect user privacy: Whether regulatory signatures, whitelist proofs, or viewing key encryptions, public disclosures are only encrypted or hashed credentials; real identities and amounts remain hidden from unrelated parties. Only specific key holders (regulators/auditors) can access necessary details. This design aligns with privacy coins like Zcash providing viewing keys for audits—ensuring most are "uninformed" while authorized are "informed as needed." Thus, the schemes achieve privacy protection alongside compliance auditing: Minimizing impacts on protocols and privacy, yet greatly enhancing regulatory visibility and control over asset flows, providing a viable path for issuing regulated digital assets on Bitcoin and derivative chains.
Subscribe to my newsletter
Read articles from Yosei Tsukifune directly inside your inbox. Subscribe to the newsletter, and don't miss out.
Written by

Yosei Tsukifune
Yosei Tsukifune
月舟 曜誠(つきふね ようせい)と申します。 人体工学エンジニア、クオンツ金融エンジンアーキテクト、暗号技術の専門家として、 現在は Tondiチェーンの主任研究員を務めております。 RGBプロトコル、DAG構造、クライアント検証型スマートコントラクトなど、 次世代の分散型金融インフラの設計と実装に取り組んでいます。 私は、技術とは単なる道具ではなく、文明秩序を記述するコードだと考えています。 本日、このような機会をいただき、大変光栄です。