Most ways into Web3 are a chart, a buy button and a signup form. This is the full on-chain desk: AMM curve modelling, a sandbox for your own contracts, a bot bench, a real liquidation engine and a multi-jurisdiction tax ledger — none of it behind an account or a wallet connection.
| Capability | The usual on-ramp | DexLadder |
|---|---|---|
| Account | Email, password, often KYC | None. Nothing is stored server-side |
| Fill model | Mid price, no cost | Spread, venue fees and real price impact from the pool curve |
| DeFi | Absent | Uniswap v2 and v3, Curve StableSwap, staking, IL lab |
| Your own contracts | Absent | Local fork sandbox — your RPC, node-signed transactions, arbitrary ABI console |
| Bots | Preset templates, if any | No-code builder plus bring-your-own via sandboxed JS or a local signal endpoint |
| Derivatives | Simplified leverage toy | Index-mark liquidation, maintenance margin, funding per interval |
| Execution | Market and limit | OCO, TWAP and pool routing |
| Backtesting | Separate paid product | Bar-by-bar replay with Sharpe, Sortino and the underwater curve |
| On-chain | Absent | Cross-checked pool radar and an explorer in the same terminal |
| Tax | Upsell to a third party | Four jurisdictions, computed on-device, nothing uploaded |
| Record | A number on a dashboard | Hash-chained, tamper-evident, yours |
| Price | Free tier with a funnel | Free, no ads, no affiliate exchange links, no token |
| Order book | A depth picture you cannot trade against | 40 levels a side, depth anchored to the asset's real 24h volume — your order walks it |
| MEV | Not modelled | Opt-in sandwich bots that only attack when the arithmetic pays |
Most practice environments draw a depth chart and then fill you at the mid price anyway. Here the ladder is the fill. Forty levels a side are built around the live mid, and the liquidity resting in them is anchored to the asset's own 24-hour volume rather than to a constant — so a deep major absorbs size that a microcap cannot, and the spread widens with thinness from a fraction of a basis point on the deepest pairs to tens of basis points on the thin ones.
What follows from that is the part a simulator usually skips. Your order walks the ladder and pays each level it consumes, so the average price of a large order is worse than the touch. The liquidity you take is genuinely gone and heals back over roughly six seconds, which means three identical orders in a row cost progressively more — the same way they would on a real venue. A resting limit order has queue priority: it records the size already sitting at its price and does not fill until the tape has traded through that queue, so a limit at a level the price merely touched is not a fill. An order larger than the book fills what exists and the remainder rests or cancels according to its time-in-force, and post-only rejects rather than quietly crossing the spread and paying taker fees.
None of that is decoration. Reading a ladder, sizing to the depth in it and knowing why your average fill differs from the price you clicked is the skill that separates a trader from someone pressing a button, and it cannot be learned in an environment that invents liquidity on demand.
Turn the sandwich bots on — they are off by default — and a swap on the pool venue becomes a target. The bot buys ahead of you to the worst price your own slippage tolerance still permits, you fill there, and it sells behind you. It does not do this theatrically: it attacks only when your tolerance is loose enough to be worth taking and the profit clears both legs' pool fees plus two high-priority swaps' worth of gas. Below that threshold, being sandwiched is not profitable and you are left alone, which is exactly the real condition.
The preview shows you the clean price and, separately, what happens if a bot sandwiches you — and when it does, the fill matches that prediction exactly rather than approximating it. That is the whole lesson in one place: a 3% slippage tolerance on a thin pool is not a convenience setting, it is an invitation with a price on it. Gas is modelled beside it on the same rules, with priority tiers, congestion read from live network conditions, and the three fates a transaction can actually meet — included, pending long enough to drift and revert, or dropped — with the gas paid either way.
Everything runs on your device. That is why there is no account to create, no data to breach, no upsell path and no reason to keep you engaged. It also constrains the product honestly: nothing syncs between your devices, and clearing browser storage clears your history. That trade was made on purpose.
No account means no password to leak, no email list, no breach and no re-engagement machinery. No server means no operator who can see your positions, your tax ledger or the wallet you looked up. No real money means the product has nothing to sell you and no reason to push you toward a live account — which is why there is no deposit button anywhere in it. Those three constraints are not features bolted on; they are why the rest of the product could be built the way it is.
It does not sync between your devices. It does not hold funds or execute a real order. It does not keep an unlimited archive — the sealed record is a bounded window. It does not give signals, price targets or advice. If any of those are what you need, this is the wrong tool, and saying so is cheaper for both of us than discovering it later.
No. It is a full Web3 desk: DeFi curve simulation, a local fork sandbox for your own contracts, a bot bench, perpetuals with an index-mark liquidation engine, bar-by-bar backtesting, an on-chain explorer and a multi-jurisdiction tax ledger.
Nothing. There is no paid tier, no advertising, no affiliate exchange links and no token.
Everything runs on your device, so there is nothing to sign in to. The trade-off is that nothing syncs between devices and clearing browser storage clears your history.
Open the desk — free, no account — no account, nothing to install