Bondify

[Bonna | U酪乳 @bonnazhu]

RWA 能做的不止价格敞口（Perp + 券商直买 + 代币化)
收益类敞口，如借贷、杠杆、分级、互换等远未开拓完
而目前这块最大的机会和痛点仍在即时流动性
那天和 @CIAN_protocol 的创始人 @Luffy_Cian 聊
想借此把一些实操层面的思路再做一次梳理和论证
也把非常复杂的 @Bondify_xyz 做一个拆解

--------------------

收益类RWA流动性难，难在它们中的大部分，在链下本就没有活跃的交易市场。把它们引入链上，等于要创造一个原本不存在的东西，这固然是价值，但难度可想而知。

此外，现有的 DeFi 基建，其实都是给货币市场工具、也就是超短久期产品设计的，用户也非常看重灵活性和流动性，要能随存随取。然而现实中大部分收益类 RWA 都是有比较长的申购、赎回周期的中长久期资产，直接搬过来，很容易需求错配、水土不服。

在我看来主要体现在三点：

- 大多属证券，不能无 KYC 流转，进不了无许可池
- 交易不活跃，LP池手续费低，无法满足资金机会成本
- 有较长申购、赎回期，NAV 和成交价间隐含资金成本

而大量收益类 RWA 上链是为了获取 DeFi 的可组合性，以及更便捷的杠杆，而非交易，这便成了更大的问题。

比如：走 DEX 直接 Loop，会面临池子深度不够和滑点问题。而走一级申赎，则是需要经历每一轮都要等待的痛苦，建仓和关仓时间被大幅拉长。尤其是关仓，借贷市场有 LTV 上限的约束，你每一圈能解锁出来去赎回的金额是有限的，尤其是你杠杆拉很满的情况，得一圈一圈退。

所以市场是非常需要专门服务 RWA Looping 一键进出解决方案的，关键就看什么样的产品形态。

我们从仓位进出、转让、以及清算/风控三个角度来谈：

--------------------

一、 进出：自建流动性 VS 聚合流动性

聚合流动性路线的典型就是 @3f_xyz 也是最早跑起来的，其给 Looper 提供的过桥资金来自外部的 Bridge Facilitator，而支持的底层资产和 Looping 策略则是来自Morpho。好处是业务模式轻，但坏处也是对外部的依赖性，目前暂支持建仓，还不支持关仓。

而 @MidasRWA @symbioticfi @grovedotfinance 这类虽也属自建流动性，但本质上它们的快速解锁只是帮你更快完成一轮的申购赎回，并不是为 Loop 而生，每一轮退出的即时性是解决了，但完整退出，仍需要多轮循环。

只有 @GearboxProtocol 和 CIAN 的新产品 Bondify 才算是真正为一键杠杆循环而设计的，而自建流动性的好处在于团队可以掌控想要支持的 Looping 策略，并推行一套独立的风控和定价参数。但两者也有区别：

Gearbox 是直接沿用它现有的 Lending Pool 和 Credit Account 架构，Credit Account 本身是一个隔离的智能合约，对应一个 Looping 策略，抵押品和借来的钱都锁在里面，Lending Pool 则是垫资方的资金池。建仓时直接从Lending Pool 一次性抽调足额杠杆资金，关仓时一次性赎回并偿还所有债务，均无需循环。建仓和关仓的过程也都是直接走 RWA 的NAV 面值直接 Mint/Redeem 完成，不存在二级市场滑点。

CIAN 的 Bondify，同样是自建流动性，为一键循环而设计，但它的资金层 SLF（Standing Lending Facility，常备借贷便利）是一个更精细设计的优先出借层，除了赚取给 Looper 建杠杆仓垫资的利息之外，还在上层的收益权交易市场扮演了一个"最后供给方"的角色，多赚一道收益。并且SLF的利用率会控制在 80% 左右，常备至少 20% 的 USDC 作为储备，存款人的流动性会好很多。

--------------------

二、交易：杠杆策略代币化

但 Bondify 比较有意思的地方，是在一键开关仓之上又多走了一步：把策略本身做成了可以买卖的标准化代币。

为什么要这么做? 有两个很现实的动机。

第一，一键开关仓仍旧要等一个完整的申购/赎回周期才算真正完成，对于那些申赎要 7 天甚至 30 天的资产，退出和建仓仍有时间成本，这个没法靠垫资越过。第二，当 Lending Pool (SLF) 里的资金不够了，新的 Looping 仓位开不出来，就只能直接去买别人现成的仓位。

而且不止于此。你也可以不把整个仓位全部卖掉，只把未来的收益打包折现卖出去，并等到这块定价降下来的时候再低价买回。这一层和 @pendle_fi 的 YT 是类似的，本质都是给用户一个更低成本去获得杠杆收益的方式。

所以在 Bondify 里，矿工的入场途径有三条：

- 走SLF 借钱自己开仓位
- 直接买现成的整个仓位 (Junior Token)
- 或者只买未来收益 (YT)

对应的退场途径也有三条：

- SLF unwind (把仓位拆掉)
- 卖 Junior Token (整仓打包卖掉、即时离场)
- 卖 YT (只卖未来收益、仓位留着)

1）YT：统一公式定价 + FIFO 队列

Pendle 的 YT 价格是 AMM 里由 YT 和 PT 双方博弈出来的，会随市场预期波动的市场价，而 Bondify 的 YT 价格直接按 SLF 利率来折现后报出来的，随着 SLF 利率的变化而波动，通过 FIFO 队列先挂先成交。

选择 SLF 利率来定 YT 的好处在于可以维持整个 Bondify系统内利率基准的统一，且能自我调节需求：Looping 需求火热 → SLF 利用率高 → SLF 利率涨 → YT 卖得更贵→ 抑制新需求，反过来，YT 便宜 → 吸引买家 → 新YT 消耗 SLF 额度 → 利用率回升，利率回升 → YT 回贵。

2）Junior 部分，特殊AMM去定价

Junior Token 买卖走的不是公式，而是会发生在一个叫 Staple 的特殊双边做市 DEX 上。它和普通 AMM 最大的区别，是把 Bid 和 Ask 分开：你卖拿 Bid 价，你买付 Ask 价，实时变动，中间价差是 LP 的利润，不像普通单价 AMM 报一个价，让套利者挑便宜的一边薅 LP。

它还有两个并行的报价渠道：PMM (协议自己的算法报价) 和 OTC (任何人都能挂固定 Bid/Ask 的公开报价)，类似自做市 + 外部做市商综合。

而任何一个 Junior Token 的价格，一般情况下会在两个"买家自己动手"的成本之间浮动：

上限 = NAV + 自己 Loop 建一个同样仓位的成本（溢价）
下限 = NAV − 自己 Unwind 退出要损耗的成本（折价）

成本包括时间，利息，滑点等等。高于上限，买家就自己建仓了，不来买了，而低于下限，卖家宁可自己一轮一轮退，也不会贱卖。所以正常情况下，Junior Token买是会需要付出一定溢价的，卖需要承担折价，Staple DEX 的Bid/Ask 机制正好是符合这点的。

--------------------

三、清算、风控：外部第三方清算 VS 内部瀑布结构

大部分 DeFi 借贷的清算，靠的都是外部的第三方清算人，你 LTV 越线，任何人都可以冲进来对你的仓位发起清算，在公开市场上把你的抵押品折价拍卖掉，赚走清算奖励。这套在加密原生资产上比较能 Work，因为抵押品在DEX 上有流动性，可以被砸掉。

但很多 RWA 不一样，即便是清算者强制发起抵押品的赎回，由于很多 RWA 的一轮赎回周期比较长，有的7 天、30 天才能拿到钱，这意味着清算人垫钱把债还掉之后，还要承担一段时间的资金占用和价格不确定性 (这期间抵押物还可能还在继续跌)，这种活，一般清算人可干不了，等于是承担了额外的垫资风险。

正因如此，Bondify 对这块做了额外的一些考量：

第一，不采用外部清算人，全部由协议执行。仓位接近清算阈值时，先触发预警、部分减仓，主动帮降一点杠杆，只有当仓位真正资不抵债时，才由协议整体全额清算。

第二，SLF的资金利用率最多 80%，常备 20% 储备金"过桥"垫付。清算的本质是赎回 RWA 抵押品、拿回 USDC 还给出借人，但 RWA 赎回有周期，中间有个时间差。Bondify 的做法是：清算发生的那一刻，先从 SLF 那 20% USDC 储备里拿现金，立刻把出借人的债平掉，同时在后台赎回，慢慢等拿到钱了再把现金补回储备池。

第三，真出现亏损时，按优先劣后逐级吸收。如果抵押品赎回出来的钱不够还债 (底层减值或违约)，缺口才按损失依次吸收：先是 YT 买家的预付款，再是 Looper 的超额抵押，即 Looper 的 自有资金，然后协议 Treasury 兜底，SLF 存款人永远坐在最优先位置，最后才被触及。这其实是很难得的，直接影响 SLF 的安全系数。

而除了这套协议主导的硬清算，Looping 整个仓位的可交易性，其实本身就是一种软性的去风险，持有人多了一条路，看空的人，可以不必硬抗，而是在仓位恶化、还没到清算线之前，直接把它在 Staple 上卖掉，把风险转手给愿意接的人即可。而愿意承接的投机者，可以用折价买进一个底层可能已经有一定风险的杠杆敞口。

本质就是让风险被实时定价和被市场连续消化，这是比单纯一键进出和垫资，要更完善的一种设计考量，能看到Bondify 的风控其实是两层的：

- 外层是二级市场，让风险被市场连续定价、转手消化
- 内层是协议主导的清算，作为市场消化不掉时的补充

就像船闸一样，两道闸门一前一后泄洪

[Bondify @Bondify_xyz]

Bondify Arena update ⚔️

⚡ One of the two dealers took the full PT-USDat position at $1,330 and 92.8% of the v-wmtUSDC position at $1,163. The other dealer took the remaining v-wmt tranche at $189. Each transfer settled in a single transaction, gas between $0.13 and $0.14. A 0.3% marketplace fee applies on top, paid by the buyer.

🎯 Check the table from that post. Selling PT-USDat at list was projected to give up $6.4 against equity. The fill realized $6.35, one transaction, zero unwind slippage. Building this position took 23 transactions across 11 rounds. Transferring it whole took one transaction and 14 cents.

🧾 Seller side. A recovered $2,682 in cash across three sales, above the self unwind floor on every fill. One caveat goes in the ledger. The v-wmt listings were priced off the oracle mark, which sat below redemption value after a vault reserve trigger. The mark reverted over the following two days, and that reversion belongs to the buyers now. Lesson booked: price listings off redemption rate, use the oracle for liquidation math only.

🔬 syrupUSDC found no buyer. The listing walked down on the 24h schedule, hit floor, and A unwound it alone on 7/14. That gave the model its first live test on the unwind side: 22 transactions, $3.68 in gas, $8.45 of total cost against equity. The published estimate had been $70.8, the corrected model said $21.6. One fix is in: swap fees scale with volume, so the slippage base moved from full collateral to debt. The residual gap is under review. Sell side estimates landed within cents. Unwind side estimates ran 2.5x to 8x hot. Deep liquidity assets with no exit window are simply cheap to unwind alone.

📊 Scoreboard through 7/15, all three books marked.
🅰️ A, who built and sold everything: $4,957 of $5,000, fully in cash. Round trip cost 0.9%: construction friction, gas, one self unwind, one sale priced off a stale oracle.
🥇 Dealer 1: +$61 on $2,500, holding both positions bought from A. Most of it is the stale mark converging back.
🥈 Dealer 2: +$9 on $2,500, one small tranche, rest in cash.
💡 Read the three numbers together. The dealers' combined gain roughly mirrors what A gave up beyond friction. A sale priced on a lagging mark moves money from seller to buyer, and the informed side gets paid. That is dealer flow at work.

📐 The model prices every listing between two exits. Ceiling: equity minus a one step exit. Floor: equity minus unwinding alone. The ratio of the two costs, per position: syrupUSDC 3.1x on paper, PT-USDat 9.7x on paper, v-wmtUSDC 2.6x on paper. One of the three has now met reality, and reality came in cheaper. The multiples that survive will be the ones built on frictions that do not go away: fees that repeat with volume, exit windows that take days, and marks the other side has not checked.

[Bondify @Bondify_xyz]

🟢Bondify Arena update
Last week we built three leveraged carry positions with $4,500 and set a baseline of $4,995.22, expecting roughly $10 in accounting gains by mid-week. Two things to report.

First, the carry check. PT-USDat is accruing on schedule. The PT price crawls toward par at about 2bp a day, matching its implied fixed rate. syrupUSDC's oracle moved 0.01% in four days, far below the estimated rate. Carry reaches the ledger through a different channel for each asset: price accretion for PT, oracle updates for yield-bearing tokens. Short windows mark it unevenly. The $10 target was missed.

One correction to last week's numbers. The 56% estimate for PT-USDat came from seven-day price change, which includes secondary market movement in the PT price itself. PT carry is fixed. Current price against par at maturity implies about 7.5% on the collateral, roughly 11 to 12% at the position level. Screening now uses carry to maturity for PT assets.

Second, the book moved from cost marks to oracle marks. Account value reads $4,895.32, PnL is negative $104.68. Three components. $4.78 construction gas, reported last week. $27 construction friction, meaning swap slippage, wrap loss, and supply/borrow price gaps, visible once positions are marked at oracle. $73 from v-wmtUSDC's oracle cutting the collateral price after a vault reserve trigger. DEX price and underlying NAV were unchanged, reserves have since recovered, and the mark should revert on the next oracle update.

Entry cost $32, against a portfolio producing about $2 of carry a day. Exit runs higher still. Unwinding a loop takes R rounds, each carrying slippage, interest, and price risk. Exit is the dominant cost in the whole trade. So this week adds an exit leg. A listed all three positions for whole position transfer.

Each listing sits between two computed prices. Ceiling is what a buyer nets taking the position and exiting in one step: equity minus one swap of slippage, price risk over the exit window, and gas, plus carry while holding. Floor is what A recovers unwinding alone: equity minus R rounds of slippage, interest and price risk across R windows, and gas. List at ceiling. Cut every 24h without a fill. At floor, unwind alone. Any fill above floor beats self unwinding by construction.

📔The book (equity / list / floor):
syrupUSDC 9x: $1,789 / $1,770 / $1,718
PT-USDat 9x: $1,336 / $1,330 / $1,294
v-wmtUSDC 5.3x: $1,265 / $1,253 / $1,245

Exit cost, same position, two paths. PT-USDat: $6.4 selling at list, $42.3 unwinding eleven rounds, a 6.7x gap. syrupUSDC: $18.7 against $70.8, 3.8x. v-wmtUSDC: $11.4 against $19.7, 1.7x. Unwind figures include gas. Listing gas is pending and sits in the cents.

[CIAN - Yield Layer of DeFi 🟡 @CIAN_protocol]

Today at 14:00 UTC, Bondify @Bondify_xyz is going live with @aave.

Join @Luffy_Cian and @alphaleaked for a conversation on the next chapter of onchain lending, covering Aave V4, RWA collateral, and the evolution of lending strategies.

🗓️ Tuesday, July 14
🕑 14:00 UTC
🎙️ Hosted by @bonnazhu

Tune in here 👇
https://t.co/zruhCWFI9Z

[Bondify @Bondify_xyz]

Based on the carry estimates at construction, the three positions are expected to generate roughly $10 in accounting gains by noon on July 13. The next few days will show how closely our carry estimates match actual onchain performance.

Construction has already cost $4.78. A rebalance would add the costs of exiting the old position, converting assets, and building a new one. Roughly $10 of short-term carry could easily be consumed by a single rotation.

Future Daily Logs will track each market’s underlying yield, borrow rate, carry, and portfolio value. We will also report the expected return before each rebalance, the actual execution cost, and whether the rotation improved portfolio performance.

All carry figures are estimates based on market conditions when the positions were built. They will change with asset prices, borrow rates, utilization, and liquidity.

[Bondify @Bondify_xyz]

Positive-carry opportunities move between markets. A position generating positive carry today may turn negative a few days later as borrowing costs increase.

Our idea is to place every eligible positive-carry trade into a dynamic portfolio and maintain it like an index. Capital is allocated across executable positive-carry markets. When a position loses its carry, its risk structure deteriorates, or another market offers a stronger and more stable opportunity, the portfolio rotates out of the old position and into the new one.

We allocated $5,000 to an execution account managed by one team member. The aim is to identify, build, and maintain eligible positive-carry positions across Aave and Morpho.

The carry must cover the cost of construction and rebalancing. Every rotation introduces gas, slippage, approvals, swaps, the cost of exiting the old position, and the cost of rebuilding a loop elsewhere. Rebalancing too frequently allows transaction costs to consume the accumulated carry. Rebalancing too slowly may leave capital in a market where carry has already turned negative.

[Bondify @Bondify_xyz]

Looking at the three positions together, the operational pain is already clear. syrupUSDC/RLUSD required switching between two DEX routes. PT-USDat required eleven rounds with a swap fee that could not be lowered. v-wmtUSDC was constrained by minimum mint requirements and needed extra capital to be completed in batches. None of the three positions could be built in one clean pass as initially expected.

The total gas bill is small, but it hides a large number of onchain operations and repeated interactions across multiple protocols: minting, wrapping, approving, swapping, supplying, and borrowing. Every additional round introduces more exposure to route changes, price movements, failed transactions, and parameter errors. The operator must also keep checking wallet balances and target leverage throughout the process. A future rebalance may require repeating a meaningful part of this workflow.

Maintaining the portfolio like an index is therefore more complicated than it appears in a spreadsheet. A new market may offer several additional points of carry, but the improvement can disappear once multiple transactions, extra working capital, or higher swap costs are included. The portfolio needs enough time in each positive-carry position for the accumulated return to cover the transaction cost. Rebalancing too frequently would allow the strategy to be consumed by its own execution costs.

Across all three positions, the execution account deployed $4,500 of capital, creating $35,451 of supplied collateral and $30,951 of borrowing. Position construction required 50 onchain transactions and $4.78 in total gas.

After construction, the account held $495.22 in USDC and $4,500 of positions marked at cost, for a total value of $4,995.22. Day-one PnL is therefore −$4.78, entirely explained by construction gas. The $4,995.22 account value is our first baseline. Carry and future rebalancing costs will continue to be recorded on the same ledger.

[Bondify @Bondify_xyz]

🔥 New JR markets are live on Bondify.

AVLT/USDC, USD3/USDC, and PT-USD3/USDC are now supported for JR trading.

With Bondify, loop positions can be tokenized into JR, making the whole position transferable

More assets, more tradable loop positions, more liquidity for DeFi ✨

[Bondify @Bondify_xyz]

This is why secondary position sales trade at a discount.

Buyers are not buying instantly realizable NAV. They are taking over debt, LTV limits, PT liquidity risk, withdrawal fees, borrow cost, and time risk.

When borrow rates stay high, high-LTV PT loops get expensive fast.

[Bondify @Bondify_xyz]

Borrow costs for the pt-apyUSD / USDC loop on Morpho have stayed high.

The NAV shown on the page is only collateral minus debt, not the value you can necessarily realize on exit.

[Bondify @Bondify_xyz]

For the 82% LTV example, assume $1M NAV, 3-day fast withdrawal, and 3.5% withdrawal fee.

Estimated full exit cost:
24h average borrow rate: ~35% of NAV
7d average borrow rate: ~39% of NAV
So $1M displayed NAV may exit as roughly $750k, $645k, before extra PT slippage.

[CIAN - Yield Layer of DeFi 🟡 @CIAN_protocol]

🚀Bondify is now live.

Yield-bearing RWAs should do more than sit onchain.
Today, many tokenized RWAs are accessible, but still underused inside DeFi.
Users can hold them, but it is much harder to build liquidity, looping strategies, yield markets, or flexible exit paths around them.
Bondify is built for that post-tokenization layer.

👇Start here:
https://t.co/jIojFNE5sF