Restaking and EigenLayer: A New Frontier for Shariah Screening

Someone asked us last month whether holding weETH was "the same as staking ETH, just with more yield." It isn't, and the gap between those two things is exactly where the interesting fiqh sits. Restaking and EigenLayer force a Shariah reviewer to answer a question that plain proof-of-stake never raised: what happens when you pledge the same capital as collateral to a second set of obligations you did not write and cannot easily read?

That is worth walking through carefully, because the mechanism has changed materially since slashing went live on EigenLayer mainnet in 2025. A lot of the halal-crypto commentary you'll find still describes the 2024 version of the protocol, where nothing could actually be taken from you.

What Restaking Actually Does

Start with the base layer. An Ethereum validator posts 32 ETH, proposes and attests to blocks, and earns issuance plus priority fees. If it double-signs or goes offline badly, Ethereum's own consensus rules cut into that stake. That is the whole arrangement, and most contemporary scholars who accept ETH as a valid asset treat the reward as compensation for a real service.

EigenLayer, built by Sreeram Kannan's team, adds a second commitment on top of the same collateral. You point your validator's withdrawal credentials at an EigenPod smart contract, or you deposit a liquid staking token like stETH into a strategy contract. Either way, your capital stays economically committed to Ethereum while EigenLayer gains the right to penalize it under a separate rulebook.

You then delegate to an Operator. The Operator opts into Actively Validated Services, or AVSs: EigenDA for data availability, plus oracles, bridges, coprocessors, sequencers, off-chain compute and verification services. Each AVS writes its own conditions. The Operator allocates a slice of your delegated stake to specific operator sets, and that allocated slice becomes slashable by that AVS.

So the pledge is genuinely layered. One pool of ETH, several independent parties holding a claim to seize part of it, each under rules drafted separately.

The Liquid Restaking Wrapper

Most retail money never touches EigenLayer directly. It goes through liquid restaking tokens: ether.fi's eETH and weETH, Renzo's ezETH, Kelp's rsETH, Puffer's pufETH. You deposit, you get a receipt token, the protocol picks the operators and the AVS mix for you, and you can go use that receipt as collateral somewhere else.

That wrapper adds two things a Shariah reviewer has to price in. First, complete opacity about which slashing conditions your money is actually exposed to. Second, secondary-market risk on the receipt itself, which is not theoretical. ezETH briefly broke from its reference price in April 2024 and cascaded liquidations through leveraged positions on other protocols.

The Reward Stack, Layer by Layer

Pull the yield apart and you get four distinct sources, each with a different fiqh character.

Base consensus rewards. Issuance and fees for validating Ethereum. Cleanest layer of the four. Mufti Faraz Adam of Amanah Advisors has argued that solo staking can be structured as a ju'alah, a promise of reward for achieving a defined outcome, and ju'alah tolerates uncertainty in the work required. Others frame delegated staking as wakalah with an agency fee, or as ijarah on computing capacity.

AVS rewards. Payment from a service that is buying economic security from you. Typically a low single-digit percentage added on top, and sometimes paid in the AVS's own token rather than in ETH. This is the layer that most resembles a real ijarah: you are leasing out the deterrent value of your capital, and the lessee pays rent.

Points. The LRT protocols and EigenLayer itself have run points programs, where activity accrues a score that may or may not convert into a future airdrop. Renzo's ezPoints fed into the REZ distribution, for example.

DeFi recycling. Taking the LRT, lending it out, borrowing against it, looping the position.

That fourth layer is where the ruling stops being contested. If you are borrowing stablecoins at a stated interest rate against ezETH so you can buy more ezETH, you have entered a riba al-nasiah transaction, and no amount of protocol-level elegance changes that. Quran 2:275 to 279 is direct text, not inference, and the loop fails immediately.

Slashing Is Live, and It Is the Real Fiqh Problem

Most halal-crypto content hasn't caught up to this part. Ethereum's own slashing is narrow and objective: double-sign or surround-vote and the protocol can prove it from on-chain data. AVS slashing is different by design. EigenLayer's documentation is explicit that slashing conditions do not have to be objectively attributable on-chain, and AVSs are free to define penalties for essentially any behavior they specify, with penalties running from a fraction of a percent up to the full allocated stake. The protocol encourages transparency and process around this rather than mandating provability.

Two classical problems land here at once.

Gharar in the penalty exposure. The prohibition of excessive uncertainty in a contract is well-established doctrine, grounded in the hadith forbidding the gharar sale reported in Sahih Muslim. When a retail holder buys an LRT, the set of conditions under which their principal can be confiscated is unknown to them and also changes over time, since the LRT's operators keep opting into new AVSs. So the uncertainty is live and ongoing, and it attaches to the corpus of the investment rather than only to its return.

Financial penalties as such. There is a genuine, old disagreement about ta'zir bi'l-mal, punishment by taking money. The dominant Hanafi position rejects financial penalties outright. Maliki and some Hanbali scholarship permits them in defined circumstances. Contemporary Islamic finance practice has largely followed a middle route: AAOIFI's treatment of late-payment penalties requires that any amount taken from a defaulting party be channelled to charity rather than enriching the counterparty, precisely so the penalty doesn't become disguised gain.

EigenLayer's redistributable slashing cuts against that principle. Slashed stake can be routed to a designated recipient address specified by the AVS rather than burned. So a service can be made whole, or better than whole, out of another party's confiscated capital, under rules that need not be objectively provable. If you find AAOIFI's reasoning persuasive, redistributable slashing is harder to accept than burn-only slashing, and the two are not distinguished anywhere in an LRT's marketing.

Where Points Fit

Ju'alah permits uncertainty in the work but requires the reward to be known. A points balance with no announced conversion ratio, no confirmed token and no guaranteed distribution is not a known reward. The workable position is that points are tabarru', a voluntary gift the protocol may or may not make, which is fine as a windfall. The problem is when points are the actual reason for the position. Building an investment thesis on an undefined future giveaway pushes the whole thing toward maysir, and you should treat expected points value as zero when you size the trade.

The Ruling: What's Settled and What's Reasoned

Be honest about which layer of the argument you're standing on.

Settled text (doctrine). The prohibition of riba in Quran 2:275 to 279. The prohibition of gharar sales. The obligation to avoid the doubtful, from the hadith of al-Nu'man ibn Bashir in Bukhari and Muslim, where the Prophet described the lawful as clear, the unlawful as clear, and doubtful matters in between that a person guards their religion by leaving.

Contested prerequisite. Whether ETH is mal mutaqawwam, valid property, at all. Mufti Taqi Usmani and the Darul Uloom Karachi position hold that cryptocurrencies lack the characteristics of mal and are not valid property, which ends the analysis before restaking is even reached. The Shariah Advisory Council of the Securities Commission Malaysia resolved in 2020 that digital assets can be treated as 'urud, tradable property, opening the door to everything downstream. Bahrain's Shariyah Review Bureau and Amanie Advisors have certified specific tokens and staking arrangements case by case. There is no AAOIFI standard resolving this, and pretending otherwise would be dishonest.

Reasoned application (inference). Everything specific to restaking. No classical authority ruled on AVS operator sets. Applying gharar and ta'zir bi'l-mal to redistributable slashing is ijtihad, and reasonable scholars will land differently on it.

Our read, offered as analysis rather than as a fatwa: base staking on a token you already accept sits in the permissible range. Restaking through EigenLayer sits in the mushtabih, the doubtful zone, and the doubt is concentrated in the slashing rulebook rather than in the reward. Any leveraged LRT loop funded by interest-bearing borrowing is impermissible on settled text, with nothing left to weigh.

Where the Other Faith Frameworks Land

The Islamic lens is the demanding one here, because the other traditions mostly don't have a hook for this mechanism.

Under Christian Biblically Responsible Investing, the six standard exclusion categories (abortion, alcohol, tobacco, gambling, pornography and anti-family content) don't touch a data availability layer. What BRI reviewers do raise is the stewardship question in Proverbs about hasty gain, which reads onto points farming and leveraged loops rather than onto restaking itself.

The USCCB socially responsible investment guidelines govern the Conference's own portfolio and address protection of human life, human dignity, economic justice and environmental stewardship. They contain no digital-asset provision, and applying them to an AVS requires an argument the guidelines themselves don't supply.

Halakhic review has the closest analogue. Ribbis attaches to loans, and staking rewards are payment for service rather than for the use of borrowed money, so the base layer usually clears. Bais HaVaad and similar authorities have been more cautious about DeFi lending against a receipt token, where a heter iska structure is genuinely the relevant tool and simply does not exist on-chain.

LDS counsel is the shortest path to a conclusion. Dallin H. Oaks warned in 1971 against speculation in the guise of investment, and Church guidance has consistently discouraged debt taken on to chase return. A leveraged points farm on an LRT is exactly the pattern that counsel describes.

What to Actually Do About It

Concrete steps, in rough order of how much they matter.

1. Separate the layers before you evaluate anything. Base staking, restaking, and leveraged LRT looping are three different rulings, and most people conflate them.
2. If you hold an LRT, find out whether it is exposed to redistributable slashing or burn-only slashing. Protocols publish their operator and AVS allocations. It takes ten minutes and it is the single most decision-relevant fact.
3. Cap the exposure. If you accept restaking as doubtful rather than clearly permissible, size it like a doubtful position, not like a core holding.
4. Purify what you cannot justify. If part of the return traces to an AVS paying in a token whose own business you would screen out, or to interest earned incidentally in a DeFi leg, calculate that share and give it away without expecting reward.
5. Refuse the leverage. Every interest-bearing borrow against your LRT is a clean prohibition, and it is also where nearly all the liquidation risk lives.

How FaithScreener Handles Restaking

Our crypto screening module covers more than 3,300 tokens and evaluates them on the token's own economics rather than on the sector label. For a restaking asset, that means looking at the reward source, the penalty design, the governance concentration and whether the position requires interest-bearing borrowing to work as marketed. A protocol whose yield depends on lending markets screens differently from one whose yield comes from selling security services, even when the two sit side by side in the same category.

Because the underlying question of whether ETH qualifies as property is genuinely contested, our framework selector lets you pick which school's premises you want applied instead of forcing a single house view onto every user. The thresholds, the sourcing and the way we handle contested rulings are all written up in the screening methodology.

The Bottom Line

Restaking and EigenLayer do not create a riba problem at the base layer. They create a gharar problem in the penalty layer, and redistributable slashing (funds seized from one party and routed to another, under conditions that need not be provable on-chain) is the specific feature that pushes a reasonably clean staking position into the doubtful zone. The one thing to remember: your reward comes from the AVS, and so does your risk of confiscation, so read the AVS list before you read the APY.

This is educational research rather than a religious ruling or personalized investment advice, and you should confirm your own position with a qualified scholar or advisor.