dan retweeted
The @AxiomExchange trading platform on @solana has just raised its fees to +1.60% on all SOL <> USDC trades.
You might think that Axiom has a flat fee of 1%, but when you take a closer look at the trades, you realize that they use their own propAMMs, which charge fees 0.60% higher in average than traditional AMMs.
When a token is designed or created to be traded in USDC and you're trading in SOL, you have to exchange that SOL for USDC and then buy your token.
Axiom doesn't give you a choice.
You're forced to go through their propAMMs and end up paying between 0.6% and 1.5% fees, depending on the comparison (traditional AMMs or other propAMMs).
For every 11.88 SOL traded at that time, instead of receiving 995.72 USDC, you receive 989.36 USDC.
+6.36 USDC for Axiom
-6.36 USDC for the trader
This is what I discovered while using Circular Replay, which we just launched.
It allows you to replay transactions with their state at the time of execution, enabling you to simulate the best quotes during execution.
We've also released a Git repository to help you use it: github.com/Circular-fi/circu…
But between us, the best thing to do is feed the repository to your AI and start rediscovering the chain.
tx: 4fckb2Lpjvj4jsRFQBp7P2TqJPiTPtNMac28EkMbExPRA6n9KC68eJ5virgio1ZVGTvuU96Cd7nP7dZvPjeuFXtn
REPLAY.
Every searcher or protocol on @solana has lived this: a transaction fails, you simulate it... and it passes.
Your RPC gives you today's state. Your tx ran against something else: different accounts, sometimes different program versions.
That's why I'm proud to announce Circular Replay today.
Giving our users the exact state any transaction ran against, rebuilt in its slot and replayable locally in LiteSVM, is huge.
Searchers/HFT, Traders/Quant, Market-makers, Protocol teams... no more guessing why you lost a trade.
More tools are coming.
More data is coming.
This is just the beginning.
Replay.
100,000 USDC > JitoSOL, split across 6 venues and 6 pools by @0xProject. Live on my #Breakpoint badge
dan retweeted
🆕 updated our price impact ux
price impact increases as your trade uses more available liquidity, resulting in worse prices
price impact is not slippage!
our updated ux makes the difference clearer
dan retweeted
Sex is great, but have you tried replacing all your nasty lookup table code with transaction v1? Incredible.
dan retweeted
We put in a phenomenal amount of work to get hooks to work correctly in 0x API, but fundamentally they set bad expectations between hook authors and aggregators. Now we have the data to back up just how bad the problem is.
nitter.cf/duncancmt/status/20335…
Just getting started 🏎️
dan retweeted
0x Cross-Chain API is generally available with 12+ bridging partners integrated on day one.
Here's how they showed up 👇
Access to historical account state on demand on Solana is a superpower
Termina is a historical simulation platform for Solana, to answer a question: what if?
Instead of guessing how new quoting logic, an upgrade, or an execution tweak will hold up, teams replay it against mainnet history — the same volatility, the same flow, the same edge cases. What used to be a deployment risk becomes a question with an answer.
Builders across DeFi — PropAMMs, DEX aggregators, lending — are already running their changes through it.
You can find the full documentation here: docs.termina.technology/docu…
dan retweeted
Glad to add another stellar router to the @Titan_Exchange Meta Aggregator. Connecting to the best algorithms to provide traders the outperformance they deserve.
We have added @0xProject's Settler as a router on Titan.
Our users can now access Settler routing as part of Titan's meta-aggregation.
Simulating pAMM outputs is non-trivial. If you're checking quotes without varying these params, your sim may be incomplete
Swap output depends on much more than pool state, reserves, and input size:
- outer program (router)
- router method
- price/CU (*)
- presence of extra accounts (**)
- presence of other swaps
and much more...
(*) some pAMMs check the priority fee from the ix sysvar and adjusts bps
(**) a pAMM that doesn't want to be bundled adds a surcharge (historically 1bps) if you don't include a specific vote account in the passed accounts
reporting from the trenches of the aggregator wars:
>$800m simulated volume on @matchametaxyz in feb 2026, split by aggregator
“simulated” = we don’t show raw quotes, we simulate execution to adjust overquoting and avoid reverts
ethereum first, arb, base, monad in thread below
dan retweeted
This will be a bit spicy
- PropAMMs have taken over Solana DEX volumes and are gaining traction in the EVM ecosystem.
- While propAMMs may become a revolutionary trading primitive that replace the conventional, less efficient AMMs, today these systems are ripe for abuse.
- This talk presents our findings of propAMMs engaging in exploitative practices and how 0x proactively protects our customers and their users from dodgy market actors looking to exploit their order flow.
Taking the main stage at ETHDenver: @willwarren
Discussing PropAMM shenanigans and what it means for onchain markets
Add his talk to your calendar in the replies 👇
dan retweeted
Bridging native @USDC with CCTP used to require a separate claim transaction
Forwarding makes that step optional
We’ve integrated @circle’s forwarding service into @matchametaxyz
Here’s what that changes 👇
Introducing our new crosschain forwarding service: a fully managed way to automate destination-chain execution for developers
Now available for CCTP on testnet, the forwarding service:
→ Automates the final mile of crosschain transactions
→ Reduces end-to-end speed, operational costs, and third-party risk
→ Improves both developer and user experience
Our forwarding service is part of a broader effort to simplify crosschain app development with the Circle interop stack. More to come.
Try it now: developers.circle.com/cctp/c…
Learn more: circle.com/blog/introducing-…
just wait until the highload ppl find this one
Excited to share a new mechanism design challenge from me and @danrobinson
Can you build the best AMM fee strategy that thrives in a dynamic trading environment?
Give it a shot and submit at the link below
ammchallenge.com/
1/
dan retweeted
The harnesses are still not optimally utilizing the models
Everyone rightfully loves claude code right now, but it is clear to me it is underutilizing the capability of opus.
The lowest hanging fruit is /init in existing code bases. It does a good job of building it's claude.md but it neglects identifying and building out relevant skills! There are enormous unlocks available when the model can identify it will benefit from a skill, or a subagent+skill, or a skilltree, and construct it automatically as part of setup but more importantly as part of continuous iteration
the harness should be seeking opportunities for whenever it makes mistakes to patch those mistakes in its own thinking with skills and protections and reminders to research things!
This is one of the major powerlevel gaps between experienced users and the unfamiliar. I am constantly prompting claude to update all of our relevant skills and claude.md whenever we finish a task or a bug. The harness begins to form to the codebase while at the same time shaping it back, making a much more usable and effective tool
dan retweeted
IDK who needs this, but @matchametaxyz is one of the best dex-aggregators I've used in a long time.
I have zero affiliation with them, but you literally just go to the page and swap. Good routes, too.
No getting stuck in Cloudflare, no weird fee mechanisms.
Incredible tech.
dan retweeted
If you care about prices, just use matcha meta




































