Let’s learn about PEPE: New Etherscan tool uses OpenAI to read smart contracts

Ethereum blockchain explorer’s new feature lets users get help understanding solidity smart contracts

article-image

Adisak Riwkratok/Shutterstock modified by Blockworks

share

Want to learn about Ethereum smart contract functionality, but have little to no developer experience? There’s a decentralized app for that.

Etherscan, the preeminent block explorer and analytics platform for Ethereum, is beta testing a new feature — called Code Reader — a tool that makes it easy to query OpenAI’s large language model (LLM) to research solidity smart contracts.

Just as the recent Advanced Filter makes it easier to research transaction information history, Code Reader provides an intuitive way for non-developers to learn about how specific smart contracts function.

Etherscan stresses that the answers are coming from OpenAI’s LLM, not Etherscan itself, and that while the tool is known to have useful applications in writing and parsing code, it shouldn’t be relied upon in a vacuum.

Professional auditing firms are already finding it offers a useful productivity boost, such as Michael Lewellen, a security solutions architect at OpenZeppelin.

“I’ve actually found it very useful,” Lewellen told Blockworks. “It’s mostly been a UX improvement, but I’ve very much enjoyed it.”

Example use cases

Bearing in mind the caveats, Etherescan says its feature will help developers and other interested parties to better understand code. 

“This feature is particularly valuable when interacting with a smart contract for the first time, as it provides a thorough understanding of its operations and functionality,” according to a tutorial on the feature.

By entering a smart contract address, one can quickly get a list of smart contract functions, which, among other uses, can allow one to interact directly with the smart contract on the respective contract address page — an alternative to using a front-end interface.

More experienced developers can use Code Reader to learn how to integrate a given smart contract with their own decentralized apps using natural language.

Is PEPE pointless?

Code Reader won’t be able to expound upon the utility of a meme or what value any given token has to its “community,” but it can break down what’s in the smart contract code.

For a simple example, let’s see what the token contract for the recently-in-vogue memecoin PEPE does:

A quick look at the PEPE token page, shows that it has around 121,000 holders and a fully diluted market cap of some $600 million.

In this case, the contract is “ownable,” meaning an account (owner) can be “granted exclusive access to specific functions,” four specifically:

  1. A transfer ownership function allows the owner to transfer ownership of the contract to a new account.
  2. The owner can set rules for token transfers.
  3. The owner can add or remove addresses from a blacklist.
  4. A function to renounce ownership of the contract, removing any functionality that is only available to the owner.

“Renouncing ownership of a contract means giving up control over the contract and its functionalities,” the Code Reader elaborates. “After renouncing ownership, no one will have the authority to modify contract state or perform administrative actions.”

That begs the question, “has ownership of this contract been renounced?” And, OpenAI quickly points out, yes it has.

So, what does that mean, precisely?

“New addresses cannot be added to the blacklist after the ownership of the contract has been renounced. The `blacklist` function can only be called by the contract owner, as specified by the `onlyOwner` modifier in the `blacklist` function. Once the ownership has been renounced, there is no contract owner, and therefore, no one can modify the blacklist. Therefore, the contract will not have the ability to limit transfers by adding addresses to the blacklist.”

Well, that’s comforting! So, you can freely send PEPE back and forth without risk of censorship (but not much else).

The feature is still under development and the Etherscan team is actively soliciting feedback. It requires an OpenAI API key — and sufficient credits — to work.


Don’t miss the next big story – join our free daily newsletter.

Tags

Upcoming Events

Hilton Metropole | 225 Edgware Rd, London

Mon - Wed, March 18 - 20, 2024

Crypto’s premier institutional conference returns to London in March 2024. The DAS: London Experience: Attend expert-led panel discussions and fireside chats Hear the latest developments regarding the crypto and digital asset regulatory environment directly from policymakers and experts.

Salt Lake City, UT

WED - FRI, OCTOBER 9 - 11, 2024

Pack your bags, anon — we’re heading west! Join us in the beautiful Salt Lake City for the third installment of Permissionless. Come for the alpha, stay for the fresh air. Permissionless III promises unforgettable panels, killer networking opportunities, and mountains […]

recent research

Research report - cover graphics (1).jpg

Research

In this report, we dive into crypto private market data to gather insights on where the future of the industry is headed. Despite a notable downturn in private raises, capital continues to infuse promising projects that aim to transform payments, banking, consumer experiences, community, and more, with 2023 being the fourth-largest year for crypto venture capital.

article-image

The stock price jump comes after Coinbase reported ending its seven-quarter run of net losses during the fourth quarter

article-image

BUZZ holds shares of Coinbase, Robinhood and MicroStrategy

article-image

Opinion: Even though I didn’t pay for my “Diamond Hands” burger with BTC, don’t let that fool you into thinking that crypto’s development is futile

article-image

The results mark “a major positive inflection point,” one analyst says, as the exchange carries net income momentum into a crypto rally

article-image

While the slate of 10 US spot bitcoin funds have tallied $4.6 billion of net inflows thus far, half of the field is lagging the leaders

article-image

Trading volumes totalled $154 billion in Q4, including $125 billion in institutional volume