# tx-sitter At [Worldcoin](https://worldcoin.org) we are looking to build an open-source transaction manager. Our motivation is to save fees compared to existing solutions, to run a high transaction volume while keeping our private keys private, and to contribute a useful project back to the ecosystem. If you are interested in being a potential collaborator or early user, we'd love if you reached out and gave us your feedback or potential feature recommendations. - Existing [tx-sitter repo](https://github.com/worldcoin/tx-sitter) - Link to tx-sitter [Telegram group](https://t.me/tx_sitter) ## Feature wish list (descending priority order) - Provide **an API for sending and managing ethereum transactions** across a set of senders - The API abstracts away details such as nonce management, transaction resubmission, and gas pricing. - Multiple senders (relayers) with different credentials (separation between apps) - Handle **transaction signing** and **key management** through [a variety of signers](https://github.com/gakonst/ethers-rs/issues/1), particularly including secure key storage in AWS KMS - Provision a **reliable upstream connection** - it’s resilient to connection failures, it reconnects with exponential backoff - connects to multiple upstream JSON-RPC providers and monitors each for currency - forwards received transactions to all upstreams - Easier **reporting** - Provides a single source of truth regarding all our blockchain interactions - Our total gas spend is a single SQL query away! - Centralizing our tx infrastructure enables **monitoring and alerting** - tracks and reports the balance of each account it manages - alters when market gas prices rise above threshold - alerts when daily/hourly spend crosses soft limit - alerts when account balance drops below threshold - alerts when transactions are stuck for longer than threshold - Provides **management api** and **spend policies** - “panic button” which halts all spend, or all spend for specific services / senders. - transactions can be approved/rejected based on configurable rules - senders can be limited to specific whitelisted receivers - accounts can be given hard spend limits - stuck transactions can be manually escalated or replaced with no-op transactions - Specify max gas price - Specify different gas price models (normal, fast, …) - Forwards **metatransactions**: - properly signed transactions which match a set of policies are given an appropriate amount of gas and sent to the chain - Advanced **sender management**: - **virtual accounts**: transactions may be sharded across multiple sending accounts for better performance (by default the geth mempool only holds ≤64 txns from each sender) - **flashbots escalation** for important stuck transactions - **transaction priotity** and reordering, preventing priority inversions for services which might experience them - Handles **multiple chains**: - initially: just EVM chains - ETH Mainnet and Testnets - Polygon & other rollups - Transaction status API - The relayer should return on submission an txId (and maybe also the intermediate txHash, which might be subject to change) - There should be an API endpoint that returns the tx status based on txId - `'pending' | 'sent' | 'submitted' | 'inmempool' | 'mined' | 'confirmed' | 'failed'` (Open Zeppelin) - Endpoint to list all tx since time delta (useful for syncing, e.g. for activity stream for wallet) - TS / Rust sdk or wrapper around ethers ## Other Specs: - Database schema: [https://github.com/worldcoin/tx-sitter/pull/6/files#diff-505bee61425c4ce3b06899c7f080d49dcb4f4963af2632860a98c13deafcd01b](https://github.com/worldcoin/tx-sitter/pull/6/files#diff-505bee61425c4ce3b06899c7f080d49dcb4f4963af2632860a98c13deafcd01b) - gRPC API: [https://github.com/worldcoin/tx-sitter/pull/14](https://github.com/worldcoin/tx-sitter/pull/14)
Last changed by  
 
dcbuilder.eth
Research Engineer @ Worldcoin. regen. math. Ethereum. ZKML. Rust. AI. Hopeful about humanity. Privacy is a human right.
518

Read more

ZKML community call #3

Attendees dcbuilder (Worldcoin) Ryan Cao (Modulus Labs) Eli Ben (Geometry Research) Priya Celeste Dosa Marek

12/7/2022
ZKML community call #2

Attendees dcbuilder Jason Morton ballestar Jonatan Vishesh stephancii killua Lakshman

11/23/2022
Worldcoin EIP4844 testimonial

Intro to Worldcoin The Worldcoin project started on the premise of giving a token to every human on earth and creating a worldwide UBI project where the value created by AGI gets equitatively distributed to all people across the globe. To achieve that scale in the far future, we are working towards onboarding the first billion people to crypto and solving Sybil resistance. We need scalable infrastructure to make these services accessible for all without compromising on network economic security and decentralization. EIP4844 Motivation More than 750k people have signed up to the Worldcoin Beta App so far – with currently more than 69k weekly active users. Our infrastructure right now is mainly composed of our Privacy-Preserving Proof-of-Personhood Protocol called World Id (currently based on Semaphore) and a non-custodial Ethereum Wallet (more specifically, a Smart Contract Wallet that already sets the stage for future EIPs around account abstraction). EIP4844 introduces a new kind of transaction type to Ethereum which accepts "blobs" of data to be persisted in the beacon node for a short period of time. Currently L2s like Optimism or Starknet post fraud or validity proofs to Ethereum calldata which is limited. As of now transactions with small amounts (<$10) are infeasible, especially in periods of high network demand (more on l2fees.info). After EIP4844 they would submit these proofs to blobspace which provides more storage and a separate fee market thus making fees on L2s a lot cheaper and allowing applications like Worldcoin to benefit from full Ethereum security for all of its users. (Learn more about L2s at L2Beat). In the past, we have built a new sequencer implementation of Hubble (a project originally started by the EF PSE team), which is an payment optimized optimistic rollup using BLS signature aggregation and custom transaction encoding. Although this leads to a massive throughput boost for transfers (<400 gas per transfer), this limits users to simple payments. However, they should be able to access the entire space of web3 applications and buy, sell, send and receive any token.

10/29/2022
Read more from dcbuilder.eth
Published on HackMD