Researcher | DeFi TG https://teapawt.org/t.co/t2UolCbKFj
Joined July 2022
- Tweets50.3K
- Following1.4K
- Followers5.4K
- Likes62.5K
Capy Research retweeted
7 GUD READS 📚 (Edition No. 113)
It's been a rollercoaster week in crypto:
> What it takes for $100M in $ENA buybacks
> Recognition of @tradexyz from tradfi
> Binance faces a $6.3B+ exploit
> Questioning the winner of the crypto options supercycle ( $HYPE and $LIT in play)
It's time to dig in 👇
Capy Research retweeted
Some of crypto’s strangest hacks ended with the money coming back.
Four cases worth remembering:
1. Poly Network (2021)
Over $600M stolen, then returned. The attacker communicated through onchain messages and even refused the bounty offered by the team.
A very unusual ending to one of DeFi’s biggest exploits.
2. Euler Finance (2023)
A roughly $197M exploit turned into weeks of negotiations and an apology from the attacker. Euler eventually announced that all recoverable funds had been returned.
This involved Euler V1, before today’s V2 architecture.
3. Munchables (2024)
Around $62.5M was taken from the Blast-based game. The developer responsible subsequently handed over the keys controlling the funds, allowing their recovery without a ransom.
4. ZKsync (2025)
Around $5M in unclaimed airdrop tokens was affected by a compromised admin key.
The Security Council offered a 10% bounty for returning 90% within 72 hours. The attacker complied, and the case was closed.
Sometimes the most important transaction comes after the exploit.
Capy Research retweeted
Growing an audience on CT is harder than ever because content has become a commodity.
A lot of people are confused by this, but it's basically impossible to grow off content alone anymore, because even the biggest ret*rds can use AI to write research threads now, it requires no skill whatsoever. Look at how many people posted reports on the Binance hack yesterday, majority of them doesn't even know what Binance is.
So what's left? Personality. That's what is very hard to imitate and the only avenue you have left to grow on CT in 2026 and beyond
Capy Research retweeted
What have @Aave’s challengers actually built?
Each takes a different approach to organizing lending markets and putting capital to work.
1. @Morpho: isolated markets and curated vaults
Morpho Blue pairs one collateral asset with one loan asset per market. Anyone can create a market, while curated vaults allocate deposits across selected markets.
Depositors get access through one vault, but their exposure depends on which markets that vault funds.
2. @eulerfinance: configurable lending infrastructure
Euler’s Vault Kit lets builders create lending vaults with their own collateral relationships, oracles and borrowing limits. Vaults can connect into isolated pairs or broader markets.
This gives asset issuers and curators more control. Users inherit the risks of the vault’s configuration and connected collateral.
3. @0xfluid: lending meets trading liquidity
Fluid connects lending and its DEX through a shared liquidity layer. Smart Collateral and Smart Debt let eligible positions also generate swap fees.
Those fees can offset borrowing costs, while positions remain exposed to changing asset composition and liquidation.
For depositors and borrowers, the specific market or vault can matter as much as the protocol name.
Capy Research retweeted
will @Enter_Elysium become the next Robinhood Chain, or lose momentum after launch like Arc?
frankly to say i lean closer to Robinhood
but i’d put the chance of comparable success at 30-40%
that’s my subjective estimate for sustained liquidity, trading volume and fees
a smaller, useful chain is also a credible outcome
here’s why i’m paying attention 👇
1) start with demand: who needs this chain, and why would they keep using it?
Robinhood has a direct route to users through its app and wallet
its chain currently holds around $1.13B in DeFi TVL
- $4.59B in DEX volume over seven days
- that gives Elysium a demanding benchmark
i’d strongly disagree if someone calls Arc a dead chain
- it still has ~$617M TVL and $1.1B in stablecoins
- its 7-day DEX volume is ~$210M
2) that raises a more useful question: how much deposited capital becomes recurring activity?
Elysium has a specific reason to exist
- built by @Kinetiq_xyz
- an Arbitrum Orbit L2 that settles to HyperEVM + uses $HYPE for gas
- the advertised targets are 100ms blocks + 300 Mgas/s
its intended users include MMs, spot trading apps & token launch platforms that need more execution capacity
i see a credible opportunity here because @HyperliquidX already attracts traders
Elysium needs to give those traders useful products and better execution
Kinetiq’s existing kHYPE business gives it relationships and an established presence in the eco
i still wouldn’t assume staking deposits will turn into active trading liquidity
3) the token launch design
the plan lets projects launch and build liquidity on Elysium
then potentially establish HyperCore spot markets and HIP-3 perps
if that works, Elysium could create additional trading activity across Hyperliquid
the planned economics also matter:
→ 50% of sequencer fees for kinetiq:native buybacks and burns
→ 25% for builders
→ 25% for operations
i’ll judge the buybacks by their actual dollar value
a large allocation means little if fee revenue stays small
4) why only 30-40%?
Elysium still needs to prove demand on mainnet
its audience is narrower than Robinhood’s, and more blockspace alone won’t make users move
there’s also an important timing detail: mainnet is scheduled for November 9
while native HyperCore market-data reads and trading access are planned for an upgrade about 4 weeks later
part of the core thesis will remain untested at launch
after launch, i’ll watch:
→ traders returning week after week
→ liquidity and volume across several apps
→ projects establishing viable HyperCore markets
→ sequencer revenue and executed KNTQ buybacks
i’m cautiously optimistic
Elysium addresses a clear need within an existing trading eco
expect there will be
i) those degens who flood in just to exploit early trends
ii) and those bad devs who come to farm degens
my conviction goes higher when users consistently pay to use it
Capy Research retweeted
Papertrade traders now have another way to trade and mine PAPER, including on Android.
@ManicTrade has launched PaperManiac, a new frontend for the same Papertrade markets, contracts and liquidity.
Alongside existing PAPER mining, it adds:
- 50% Lossback: eligible realized trading losses earn Manic Bonus Credits.
- A $2,000 USDC prize pool in November’s trading tournament.
- Mining analytics to track mint rates, estimated rewards and mining costs.
The important detail is how Lossback works.
Lose $20 on an eligible position and receive $10 in Bonus Credits. These are promotional credits, with USDC unlocked through further real-money trading on Manic: every $50 in volume unlocks $1.
Credits cannot be withdrawn directly or used to trade Papertrade.
For existing Papertrade users, the useful additions are mobile access, mining visibility and extra reward campaigns in one interface. Trading still carries the risk of losing your full position.
app.manic.trade/papertrade
I quite like Manic’s interface. It feels smooth to use.
Introducing PaperManiac 🟠
We’re bringing @papertrade_xyz to mobile with the first dedicated Papertrade App on Google Play.
Same markets, same PAPER mining, with more rewards through Manic.
Live on Web & Android:
app.manic.trade/papertrade
Capy Research retweeted
BREAKING: World's largest crypto exchange Binance hacked with over $5.6 billion stolen:
• $1.45 billion $BTC
• $875 million $ETH
• $234 million $TRX
• $50 million $XAUT
• $1.57 billion $USDT
• $823 million $USDC
• $75 million $WBETH
• $88.5 million $RLUSD
+ $447.5 million additional cryptocurrencies
This is the largest hack in crypto history.
Capy Research retweeted
Starknet, RAILGUN and Aztec take three different routes to private DeFi.
The differences matter when you ask where your assets sit, what becomes public and who can inspect your records.
@Starknet / STRK20
Adds shielded balances, private transfers and supported private swaps to the Starknet ecosystem, using existing liquidity.
Its disclosure model is built in: deposits are screened, viewing keys are mandatory, and Financial Privacy Inc operates infrastructure for targeted disclosure under lawful requests.
@RAILGUN_Project
Adds a private-wallet layer to existing EVM chains. Users shield assets into 0zk balances and access supported DeFi integrations without moving to a separate privacy chain.
Private Proofs of Innocence checks funds against known bad-activity lists without giving list providers access to private records. Users can separately share viewing keys for audits.
@aztecnetwork
Builds privacy into smart-contract execution. Developers combine private state and functions, executed on users’ devices, with public state and functions.
Privacy therefore depends on application design. Public calls and transaction metadata can remain visible. Its mainnet Alpha is live, but the official docs describe it as early, unaudited software.
For users, the choice depends on the workflow: which assets and apps are supported, what leaks at public boundaries, and who holds the power to disclose the rest.
Capy Research retweeted
alpha: these DeFi protocols help you make more trades and more LP profit from the same $1,000
a DEX can grow volume while its LPs lose money
then i found better measures of DeFi efficiency than TVL
» @1inch aqua is trying to increase the utilization of liquidity
» @Uniswap v4 hooks are increasingly trying to improve the economics of liquidity
those are different optimization problems, and several other projects add useful approaches
→ @0xfluid: combine lending and trading
smart collateral + smart debt let collateral & debt positions serve as DEX liquidity
trading fees can add income or offset borrowing costs
→ @CurveFinance: optimize for the assets being traded
stableswap concentrates depth around the expected relationship between correlated assets
→ @BreederDodo: adjust pricing around inventory
its PMM model concentrates liquidity around a reference price & changes quotes as inventory moves away from its target
this makes inventory management another part of the efficiency question
→ @CoWSwap AMM: address arbitrage losses through batch trading
its design uses solver competition to capture surplus for the pool
the relevant test is whether that improves LP returns after accounting for execution and costs
execution quality means the price users receive, gas costs and whether trades successfully fill
adverse selection means the losses LPs take when better-informed traders trade against outdated prices
i’d start with Aqua, Uniswap v3/v4, Curve and Fluid, comparing matched assets over the same period
these DeFi convince me better execution for traders and better returns for LPs from the same underlying capital
» $UNI $1INCH $FLUID $CRV $DODO $COW
Capy Research retweeted
Linn's Leverage #1139 is out: (Link in tweet below)
@boros_fi
@pendle_fi
@Chainflip
@mezzanine_fi
@Mantle_Official
@StaticsProtocol
@GammaSwapLabs
Like/Repost to support:
Capy Research retweeted
TxFlow is Hyperliquid from Temu
They forked Hyperliquid's Arbitrum bridge and changed the name. They copied HyperUnit too and forgot to change the name: txrelay[.]com literally says "Unit is the asset tokenization layer on TxRelay". The frontend, the API and the fees are Hyperliquid's. Even the slogan. The L1 is closed source, with Hyperliquid's naming on everything: TxCore, Tx-BFT, TIPs
Hyperliquid is closed source too, but it has 27 validators that each post prices, and anyone can run a node to check them. TxFlow has 5 validators, all its own, nothing you can run, and one key that acts as the entire oracle. The same key sits on the bridge multisig
So it's a centralized exchange secured by a 3/4 multisig. Their own audit says those keys "can propose arbitrary withdrawals"
And the growth plan is KOLs shilling ref codes
Capy Research retweeted
Why does @infinifi offer higher yield to users willing to wait?
The mechanism combines income from underlying strategies with a larger share of returns for committed capital.
The basic flow:
• Deposit USDC to mint iUSD, the non-interest-bearing receipt.
• Stake iUSD into siUSD for liquid yield.
• Choose a locked position for higher-weighted returns and a slower exit.
infiniFi allocates capital across liquid and longer-duration strategies. Depositors’ withdrawal commitments help it match those investments to expected outflows. Some capital backing liquid positions can also enter longer-duration strategies.
An important detail: the unbonding period starts when you request to unwind, rather than counting down automatically from your deposit.
But the locked position also carries first-loss exposure.
According to its slashing docs, realized portfolio losses hit Locked-iUSD before senior positions. Starting to unwind does not remove that exposure.
If the junior capital is exhausted, losses can reach senior depositors too. And liquid positions still depend on available reserves; direct redemptions can queue during liquidity stress.
The extra yield therefore compensates for both reduced exit flexibility and a different place in the loss hierarchy.
Capy Research retweeted
Aptos, Sui, Sei: what have @Solana’s challengers actually built?
The speed comparison only tells part of the story. Each has taken a different route toward attracting users and capital.
1. @Aptos: stablecoins and payments
Aptos supports native USDT and USDC, with payment integrations including Bitso and Yellow Card. These give users routes between local currencies and onchain dollars.
Distribution matters here: cheap transfers become useful when people can actually access them.
2. @SuiNetwork : shared trading infrastructure
DeepBook provides an onchain order book that applications can integrate, rather than each building liquidity infrastructure from scratch.
That gives Sui a concrete product beyond faster execution. Its usefulness still depends on market depth and trading activity.
3. @SeiNetwork : access for Ethereum developers
Sei’s parallelized EVM lets developers reuse familiar contracts and tools. Its first Eidos storage upgrades reached mainnet in August 2026, while further components remain planned.
Easier deployment lowers the barrier to building. It doesn’t automatically bring users or liquidity.
The capital gap remains substantial.
DefiLlama’s tracked stablecoin supply, October 26:
• Solana: $15.6B
• Aptos: $867M
• Sui: $431M
• Sei: $43M
These balances measure capital present, not payment volume or user retention.
All three have built beyond the speed pitch. The harder task is making their applications useful enough for capital to stay.
Capy Research retweeted
Nillion’s move to Ethereum L1 brings encrypted instructions closer to the applications that can use them. The trade-off is operating cost.
Following my earlier @nillion breakdown, this is a second migration: Nillion moved from Cosmos to an Ethereum L2 earlier this year. Blacklight is now moving from that L2 directly onto Ethereum mainnet.
The practical benefit is composability. Once a Covenant’s instruction is reconstructed, an Ethereum contract can act on it in the same transaction. Registration, staking, fee accounting and reconstruction also sit on L1.
What has actually been delivered?
Dusk was announced live on October 6. Mainnet contracts are published, and onchain activity includes trigger posting, share publication and successful result posting.
At my October 26 check, the docs-listed TriggerMarket showed 74 transactions. That establishes activity at this deployment, but does not identify independent customers or prove substantial commercial demand.
The costs remain meaningful. Operators need NIL stake, ETH for gas and supporting infrastructure. Small, frequent triggers may be harder to justify economically.
Ethereum also doesn’t replace the privacy assumptions: enough colluding committee nodes can reveal a payload early, and its contents become public after release.
The economic distinction matters: NIL emissions support availability; customer-funded ETH fees pay for work. The business case depends on paid workloads covering operating costs, beyond the incentives used to keep nodes available.
.@Nillion is extending its privacy stack into onchain markets.
Its products cover several different jobs:
- nilDB stores encrypted data shares across nodes, so one storage node cannot reconstruct the original data alone.
- nilCC runs application workloads inside trusted execution environments, with attestation to check the execution environment.
- nilAI offers private model inference through OpenAI-compatible APIs built on nilCC.
On October 6, Nillion announced Dusk live on Ethereum mainnet. It introduces Covenants: instructions kept encrypted until an attached condition is met.
Blacklight node committees help resolve those Covenants. Enough shares must be published before an instruction can be reconstructed and verified on Ethereum.
NIL also has roles across the stack: nilDB and nilCC use NIL-based credits, while Blacklight operators stake NIL. Their reward model includes trigger fees.
Together, these products address private storage, confidential processing and condition-based disclosure of market instructions.
For builders, the value is being able to choose which parts of an application need privacy. The layers have different security assumptions, and multichain Covenants remain roadmap work.
Capy Research retweeted
I think we need to go a little deeper on why Sorsa score still matters
it was designed to score X accounts. social media accounts. not onchain reputation, not the team behind the project, not where they’re based
Sorsa score only shows the “value” of a specific account within its niche
more smart followers, projects and funds = higher score
that’s it. Sorsa score can’t tell you whether a project will rug you onchain
TROVE’s score is high because a lot of “credible” accounts follow it
but zoom out and look at who actually follows it. can you really trust them? there's also a question to accounts still following scammers..
tldr: social score ≠ onchain reputation
as per @SorsaApp i can trust ‘TROVE’
which scammed/ran away with $9.4m from people this year from its ico, (we all remember this)
that tells you this sorsa score is all shitty AI slop score and should not be taken seriously
i checked for many scams and rugs and it told me they are credible enough
Capy Research retweeted
Staking BTC on @Starknet earns you STRK. That distinction matters more than the headline APR.
The mechanism is straightforward:
> Stake supported BTC wrappers on Starknet.
> Earn newly minted STRK.
> BTC stakers receive 25% of total staking rewards under the documented model.
For BTC holders, this adds a staking use case alongside Starknet’s DeFi integrations.
For STRK holders, attracting that BTC capital comes with additional token issuance.
Because rewards are paid in STRK, their value relative to BTC changes with the STRK/BTC exchange rate. A higher quoted APR can reflect a stronger reward token, even without more underlying activity.
An October 15 community proposal suggests making the BTC reward allocation variable to reduce those swings. That remains a proposal, not an implemented change.
The useful question is how much lasting BTC liquidity Starknet can attract for the STRK incentives it distributes.
Capy Research retweeted
Early crypto cards 💳
Recently launched / already working:
- @vPay_Global - OmniBank / AI app, virtual and physical Visa, 0% FX, cashback up to 1.5%.
- @Karta_Personal - self-custody + Visa Signature + USD/EUR accounts. Virtual card 5 USDT, 1.5% payment fee.
- @cashi - virtual Visa for stables, public beta since August, recently added XRP.
Waitlist / coming soon:
@DptPay Physical - physical Visa, waitlist open, shipping mid-November, 20% discount.
Always DYOR
New Crypto Cards are already here...
New crypto cards are on the way, I'm sure there's something you'll be interested in.
@fizenapp:
• Crypto neobank for travelers and global users. Spend USDT like cash worldwide.
• Backed by Tether. Full super-app: card + QR Pay (strong in SEA: Vietnam, Philippines + LATAM) + eSIM + tokenized US stocks + remittances.
• Card: Instant virtual Visa, Apple/Google Pay. Very high limits (~$1M daily). Low fees (0.5% conversion, often 0 top-up within plan limits).
• Region: Global (SEA/LATAM focus), accepts users from some restricted countries.
• Price: Adventure card from $9.99 (higher tiers available).
@OnboardGlobal:
• Global money app for receiving USD/EUR/GBP, spending, and handling stablecoins. Aimed at freelancers, remote workers, and cross-border users.
• Real named USD/EUR/GBP accounts + virtual card. Fund with local currency or USDT/USDC. Apple/Google Pay support.
• Card: Virtual Flex Card (first one free, up to 3 cards). Limits $5k daily / $20k monthly (higher available on request).
• Fees: Funding 1.25%, FX/cross-border $0.30 + 1.5%. No monthly fee.
• Region: 140+ countries (strong presence in Africa/Nigeria, Asia, and global professionals). Not available in the US.
@STEALFxyz:
• Privacy layer for everyday finance, built on Solana.
What stands out: Two layers compliant spending account (card/bank) + private wallet (powered by Umbra + Arcium). No on-chain link between them. You control the keys.
• Card: Visa card is coming (public beta already live with private transfers and gift cards). Cashback in solana:G5W6LwkLeoj6rZqBP1y3KT8k6Cz6rXGJmMNU3TArXtXw.
• Region: Global (Solana-native, privacy-first approach).
• Card not fully live yet, but you can already shield funds and spend privately on gift cards.
@purefiapp:
• Focus: High-yield savings + low-cost remittances (not a classic crypto spending card).
• What stands out: Up to 6% APY on real estate (shares in actual rental homes) held in a self-custodial wallet. Low fees on international transfers.
• Card: Available via licensed partners (on/off-ramp + cards). Main emphasis is on yield and remittances rather than everyday card spending.
• Region: 100+ countries (Asia, Africa, Europe, Americas).
• Best for: Users looking for yield on real assets + cheap cross-border transfers without traditional banking interest.
Always DYOR.
Capy Research retweeted
🚨 Urgent: Hiring Web3 KOLs & Ambassadors for YeNo!
Looking for active Web3 content creators and ambassadors for upcoming campaigns.
💰 Earn $100–$500/month in USDT/USDC.
Interested?
🔁 RT to spread the word
📩 DM me your stats from the last 3 months.
Let’s build together! 🤝
Capy Research retweeted
@grvt_io has bought back and burned 1.1M GRVT.
Its 2.0 roadmap raises a broader question: what happens to your capital between trades?
I examine how earning, collateral and trading connect, and who receives the value along the way.
🤝 Paid partnership
