When protocols start talking about revenue distribution, the rules change.
That is the frame Castle Labs used in this week’s Chronicle, arguing that crypto has spent years leaning on onchain activity metrics such as total value locked, volume, user counts, transaction counts and active addresses. In this cycle, the more pressing questions are different: whether the business is sustainable, whether activity becomes actual revenue, and who gets to keep that revenue once it shows up.
The report focused on a handful of examples: Maple’s new rules-based buyback plan, protocol fees going live on Uniswap V4, Sky’s strong revenue numbers paired with a hard-to-read balance-sheet story, Theo adding Fidelity International’s FILQ to thBILL, and OndoPerps turning tokenized equities into collateral. Castle Labs’ broader point is simple: revenue alone is not enough. It needs a clear path to value accrual.
Maple ties buybacks to revenue, but the destination of repurchased SYRUP is still unclear
Castle Labs described Maple as one of crypto’s standout recent success stories, serving mature allocators with professionally managed, permissioned and security-focused lending products. The firm said Maple just posted a record first half, with assets under management reaching $4.6 billion, up 81% year over year. Q2 revenue came in at $4.4 million, up 47% year over year, while the business itself grew 22% even as DeFi lending contracted 31%.
Q3 has also started on a strong note. Maple launched syrupUSDG, its first new Syrup asset in two years. The product brings Maple’s onchain credit engine to Global Dollar, or USDG, and Robinhood Chain, giving holders access to Maple-originated lending strategies.
syrupUSDG can also be placed in Robinhood Earn, where it offers up to 7% APY. According to the report, that yield wraps together Robinhood distribution, Morpho vault infrastructure, Steakhouse Financial curation and Maple’s institutional credit as the source of yield.
The question then becomes how SYRUP holders benefit from that expansion.
Maple recently put discretionary buybacks in place under MIP-019 and is now preparing to replace them with a rules-based framework under MIP-021. Under the proposal, buybacks would be directly linked to monthly gross revenue:
- Below $1.5 million: 10% allocated to SYRUP buybacks
- $1.5 million to $2 million: 20% allocated to buybacks
- Above $2 million: 30% allocated to buybacks
Those buybacks would take place at month-end after monthly revenue is finalized. The purchased SYRUP would then be assigned to the SYRUP Strategic Fund, or SSF, which Maple defines as operational capital for strategic growth, token liquidity, capital reserves and buybacks.
Castle Labs noted that most buyback programs are usually designed to distribute profits, retire supply or reduce circulating tokens. Maple’s route does not clearly do any of those, which makes the structure look more like a treasury-management strategy.
That issue has already surfaced on the forum. Community members have asked for repurchased SYRUP to be placed in a public reserve address and treated as non-circulating, non-voting and non-transferable unless a new governance proposal says otherwise. Castle Labs said those protections do not appear to be part of the current proposal.
As a result, the definition and future use of strategically repurchased SYRUP remain unresolved. In Castle Labs’ view, that ambiguity is likely to shape how the market judges the framework and may lead people to see it more as treasury management than as a mechanism that delivers real value accrual to SYRUP holders. The report said the issue is currently headed for snapshot voting.
Uniswap V4 turns on protocol fees, and the key question is what LPs do next
Castle Labs said Uniswap founder Hayden shared that the protocol is now collecting more than $5 million in fees per day, second only to USDT and USDC.
A large share of that growth, the report said, came from the launch of Robinhood Chain. But Castle Labs also cautioned that these figures are neither annualized nor necessarily durable. Its expectation is that Robinhood will internalize those fees soon.
Debate around UNI utility, protocol revenue and fee distribution has been running for years. Uniswap has now brought protocol fees to V4 as well.
Earlier fee-switch activation on V2 and V3 pushed many LPs toward V4, serving as an incentive for capital migration. The new issue is what happens now that V4 itself has fees turned on. Will that improve capital efficiency for Uniswap as a whole and help finish the shift away from V2 and V3, or will it create a fresh reason for liquidity to move elsewhere?
Because of its architecture, V4 needs a more flexible and customizable approach to protocol fees. Castle Labs said that is mainly due to hooks and dynamic fees, which allow the creation of multiple fee tiers.
Governance rules for V4 will therefore be defined by a “Fee Controller System,” a setup meant to let governance establish rules for different pool families through V4FeePolicy while retaining the ability to override or adjust those policies through V4FeeAdapter.
The central issue remains the same: how LPs respond.
Castle Labs wrote that earlier fee switches on V2 and V3 reduced LP take by 25%, encouraging some of them to move into V4. With V4 fees now activated, the protocol has to find out whether LPs stay or move to fee-free venues. The report said governance participants have already raised those concerns.
One proposed answer is conditional activation of the V4 fee switch, meaning fees would only be turned on when LPs are profitable. The report quoted one user as saying: “If a pool’s implied volatility persistently exceeds realized volatility, governance can take a cut without breaking LP trading. On the other hand, if RV > IV, LPs are already undercompensated. Taking 25% of their fees won’t make the protocol money. It will push LPs further into negative EV.”
Castle Labs said it will keep tracking V4 fund flows to see how that plays out in practice.
The report added that V4 fees are expected to be used for LP rewards and buyback distribution, with a split close to 5%-25% / 75%-95%. It also compared the setup with Hyperliquid buybacks.
So far, Uniswap has burned more than 6 million UNI. Castle Labs said additional statistics on UNI burns are available, and flagged the broader question of how buybacks and token-holder income interact with protocol revenue and fee distribution. Different projects, it said, handle that very differently.
Sky posts record revenue, but the accounting story is harder for holders to parse
June was a record month for Sky, according to Castle Labs. The report listed four highlights:
- A record revenue run rate of $419 million
- More than $250 million in USDS yield distribution
- Continued growth in Sky reserves
- Rising activity across the Sky Agent Network
Castle Labs said Sky is clearly generating revenue, launching products and driving additional activity through Spark, Grove and the broader Prime Agent structure. But the harder question is what those numbers mean for SKY holders.
There is a value-accrual mechanism in place. Protocol revenue from stability fees and Sky Agent performance is used to buy back SKY on the open market, and those tokens are then distributed to participants in the Sky staking engine.
Still, Sky is not a clean one-token, one-balance-sheet story. Prime Agents have their own roles, tokens, holders and economics.
Castle Labs explained that Prime Agents are specialized capital allocators inside the Sky ecosystem. Sky can fund or support them, they can owe money back to Sky, and their performance can feed into protocol profit. In theory, those profits should support SKY buybacks and staking rewards.
That structure makes the token-holder question much harder to answer cleanly. Investors can look at protocol revenue, cost of revenue, reserves, staking rewards and token-holder income. Once Prime Agents sit around the core protocol, though, the balance sheet becomes tougher to read.
Castle Labs broke the issue down this way: some value may accrue to SKY, some may stay at the agent level, some may move through sUSDS or ecosystem incentives, and some may accrue to SPK, GROVE or future agent tokens.
The report also cited a post this week by PaperImperium arguing that Sky’s financial position is difficult to interpret because Prime Agents are not treated consistently in the accounts. If it is too hard to determine what token holders are actually entitled to, Castle Labs said, the value proposition itself can weaken.
OndoPerps launches with tokenized equities as collateral
Ondo has launched OndoPerps, its perpetuals product for trading stock perpetual contracts with up to 20x leverage.
Castle Labs said the product complements Ondo’s spot equities offering and gives the company a way to compete with venues such as Trade.xyz, which it said has already captured a large share of perpetual open interest in RWA markets.
Listed markets include commodities such as oil, gold and silver; equities including Apple, Tesla, Nvidia, Microsoft, Amazon, Alphabet, Meta, Netflix, Intel, AMD, Oracle, Micron, Palantir, SpaceX, Strategy, Coinbase, Circle and Robinhood; and indexes including US100 and US500.
The distinguishing feature of the launch is that it extends Ondo’s existing stack by allowing tokenized equities to be used as collateral for 24/7 perpetual exposure to commodities, stocks or indexes.
Castle Labs compared that to one of Hyperliquid’s key advantages over other trading venues: unified margin across products, which opens the door to a wider set of strategies including hedging and delta-neutral positioning.
That means existing Ondo spot-equity holders can use those positions as collateral to hedge or lever their current exposure. Castle Labs said the perpetuals market still has a long runway and that RWA perpetual trading is likely to become one of the fastest-growing segments in the near term.
On the backend, the report said liquidity is supported by a mix of institutional market makers and user flow.
Theo adds FILQ to thBILL as tokenized funds move from issuance to utility
For the last two years, progress in RWA has mostly been measured by what got issued onchain, Castle Labs wrote. Another fund tokenized. Another asset manager onboarded. Another chain supported. The next stage is about what those assets can actually do.
That is why the report highlighted Theo’s FILQ integration.
Theo invested $20 million into Fidelity International’s tokenized U.S. dollar digital liquidity fund, FILQ, through Sygnum and added FILQ as the second institutional base asset in its existing thBILL product.
Holding the fund directly is the familiar starting point. Once the fund is onchain, though, the more interesting question is where else it can sit: inside stablecoins, Treasury products, collateral systems, lending markets, vaults or settlement flows.
Castle Labs quoted Theo GTM and Chief of Staff Evan as saying, “Most investors don’t want money market funds. They want yield, and they want it in a form they can actually use.”
Evan also said, “Holding the fund directly only gives you the return, and nothing else. Wrap it into a composable product, and the same dollar can earn and stay liquid at the same time... tokenized funds make it institutional grade, and onchain products make it useful.”
thBILL had already integrated Wellington’s tokenized Treasury fund before this. Adding FILQ brings in Fidelity International as a second institutional manager, broadens the product’s base and shows that the structure can onboard new managers without having to rebuild the product each time.
Castle Labs argued that if tokenized funds remain standalone wrappers, competition will stay concentrated around issuer brand, AUM and chain support. If those funds become inputs for more customizable onchain products, competition shifts to what additional value can be built around them.
Fidelity International framed tokenization in similar terms. The report quoted Emma Pecenicic, head of digital asset distribution at Fidelity International, as saying the firm sees tokenization as “a foundational shift in how global financial markets operate,” and that combining investment expertise with digital-native infrastructure can “bring regulated, institutional-grade onchain liquidity to markets operating around the clock.”
Castle Labs then asked what that shift should look like in practice. Theo’s answer is that direct ownership is step one because it matches how funds work today. The real value of putting the asset onchain comes after that: using it as collateral, routing it into strategies or settling instantly.
The report framed utility as the real destination for these assets. Can they sit inside Treasury products? Can they support stablecoins? Can they be used as collateral? Can they move through lending markets, vaults or structured products without becoming a compliance problem?
In that sense, FILQ joining thBILL is a small but useful example of the shift. The fund is not the final product. It is an asset on Theo’s balance sheet that strengthens the product and prepares it for growth and expansion into places that Fidelity cannot easily reach on its own.
Castle Labs said the next wave of tokenized-fund adoption is likely to be driven less by direct ownership of those institutional assets and more by how they are embedded into onchain products.
What Castle Labs is watching next
The report closed with several themes Castle Labs said it is monitoring.
Circle and agent payments
Jeremy Allaire published a post on the Circle Agent Stack. Castle Labs said the important question is whether the overlap between the agent economy and the onchain economy turns into real payment flows. That means watching whether agents actually hold balances, pay for services, rebalance wallets, settle bills or trigger transactions in USDC, and whether developers build around Circle’s stack instead of generic wallets and APIs.
Castle x Kaiko revenue report
Castle Labs said it will publish a report with Kaiko later this week on how operating blockchain businesses is changing. Chains used to rely mainly on blockspace fees, it said. Now they are being pushed to diversify into other verticals such as MEV, sequencer economics and application capture, while making sure they can capture value where it truly accrues.
The staying power of Robinhood Chain
Castle Labs said Robinhood Chain made a major impact on the onchain landscape in its first week. Shortly after launch, the firm marked it as the fifth-largest chain by 24-hour DEX volume, with daily volume above $370 million and cumulative volume of $1.35 billion, driven mainly by CASHCAT. It later rose to second place in 24-hour DEX volume, behind only Solana.
According to the report, Robinhood has all the ingredients to become a serious distribution channel for onchain finance, but it is currently leaning into the meme side of the market, which has helped produce strong early metrics. Castle Labs said it will be watching how those two directions find a better balance over the coming months.
SEC roundtable on IPO access
Castle Labs also pointed to the SEC’s virtual roundtable on IPO modernization and expanded access to public markets. The report said crypto is already starting to affect traditional financial markets more directly as large trading venues open up, more assets are tokenized and even IPOs begin to appear onchain.