4 named authorships.
3 merged. ERC-8203 in review.
I'm Adam Boudjemaa, a named author on four Ethereum specifications: named author (1 of 5) of ERC-3643 (T-REX, Final), first author of ERC-6960 (Dual Layer Token), and co-author of ERC-7410. I am also a co-author of ERC-8203 (Agent Offchain Conditional Settlement), second of three, and wrote its reference implementation. That one is a draft still in EIP review and not merged. ERC-3643 is Final; ERC-6960 and ERC-7410 are published Drafts. Each card links to the source and explains my role.
ERC-8203, still a draft in review, is where my standards work meets AI agents. For the rest of that story, from agent payments to mandate checks, see the web3 x AI hub. Selected implementations and Solidity tooling are on my open-source tools page.
T-REX: Token for Regulated EXchanges
The institutional-grade security token standard. T-REX enables compliant issuance and management of security tokens with on-chain identity verification and transfer compliance. Used by Tokeny and adopted by institutions managing billions in tokenized assets.
The $32B+ adoption figure is theirs, not mine. Source: ERC3643 Association, erc3643.org, 2025. For the fractional-ownership side of the same problem, see ERC-6960, the dual-layer token standard, and for why these assets ended up wanting their own chain, read why RWA tokenization needs its own blockchain.
Dual Layer Token
A multi-asset token standard that combines fungible and non-fungible properties in a single contract. Built for real-world asset tokenization where a single asset (like a building) needs both divisible ownership shares and unique metadata per fraction.
ERC-20 Update Allowance By Spender
An ERC-20 extension that lets a spender reduce or revoke the allowance an owner granted to it. It adds one spender-side function and remains compatible with ERC-20.
Agent Offchain Conditional Settlement
A minimal interoperability framework that lets autonomous agents settle conditionally off-chain, using typed cryptographic structures and on-chain extension interfaces, without dictating the full channel implementation. Primary author Xianrui Qin. I am the second of three named authors and built its reference implementation. Open draft, in EIP review.
Three different problems, three different standards. If you are splitting one real-world asset into tradable fractions without deploying a contract per slice, the answer is ERC-6960 and its dual-layer main-ID / sub-ID model. If your worry is an approved spender sitting on an allowance it no longer needs, the answer is ERC-7410, which lets the spender revoke its own allowance, without waiting for the owner to send a transaction. And if you are asking why regulated assets keep outgrowing general-purpose chains in the first place, that argument is laid out in why RWA tokenization needs its own blockchain.
Regulated tokenization and RWA writing
How to get an ERC proposal accepted
Scoping, implementing and discussing a standard, from my work on ERC-3643, ERC-6960 and ERC-7410.
Agent mandates on ERC-3643
Investor eligibility vs agent authority, and where a mandate check actually belongs.
Confidential balances on permissioned tokens
What freeze rules, transfer results and disclosure look like when balances are encrypted.
ERC-3643 across a Cosmos VM boundary
What escrow changes when a compliant token has to cross from one VM to another.
Agent payments, x402 and liability
x402 settles the payment. Delegation, spending policy and outcome records are still on you.
Choosing a security-token standard? Read ERC-3643 vs ERC-1400 vs CMTAT, the FAQ, or the full blog.
Need protocol design help? See services