Simulation mode · Paper trading only · Real trading disabled

Platform

An arbitrage intelligence engine, not a trading bot.

BaseArbBot.ai is built to answer one question rigorously: which price dislocations on Base would actually survive execution? Everything in the platform serves that question, and the answer is proved in simulation.

Active market sources
5
Markets in universe
7
Route × size combinations
714

Capabilities

What the engine actually does.

Multi-market scanning

The market universe is configuration, not code. Assets are added to a registry, verified by preflight, and then swept alongside everything else.

Executable pricing

DEX legs use on-chain quoter output for the actual size; CEX legs walk the order book level by level. The number used is what would be received, not a headline mid price.

Complete cost modeling

Pool fee tier, venue maker/taker tier, price impact, depth consumption, Base L2 gas and a safety buffer are all netted per candidate.

Directional route analysis

DEX→CEX, CEX→DEX and DEX→DEX are evaluated separately, because the profitable direction is rarely symmetric.

Trade-size sensitivity

Every route is tested across the full notional ladder, since impact and depth change the answer at each size.

Inventory-aware simulation

Fills are booked against two-sided simulated balances, so trades that would be impossible without rebalancing are treated as impossible.

Latency and decay modeling

Candidates are re-quoted before they are accepted, and spreads that evaporate in the interval are recorded as decayed rather than banked.

Venue health governance

Unreliable or rate-limited endpoints are demoted, backed off and reported instead of quietly polluting results.

Pipeline

From raw quote to simulated fill.

  1. 01

    Capture

    Server-side adapters read Base pools and centralized order books.

  2. 02

    Normalize

    Every venue reduced to an executable buy and sell side for a given size.

  3. 03

    Cost

    Fees, price impact, depth, gas, latency and buffer subtracted.

  4. 04

    Confirm

    The spread is re-quoted; anything that decayed is discarded.

  5. 05

    Simulate

    Two-leg paper fill booked against a reconciled inventory ledger.

Base DEX liquidityNormalized executable priceCentralized order books

Cost surface

Fourteen frictions are modeled before anything qualifies.

Each factor is recorded per opportunity with its provenance, so an operator can see which inputs were verified from a venue and which were estimated.

Executable DEX swap outputDEX price impactOrder-book depth consumptionMaker/taker fee tiersDEX pool fee tierBase L2 gasQuote freshnessConfirmation decayExecution latencySafety bufferInventory availabilityTwo-leg execution riskRebalancing burdenCapital utilisation

Inventory

Capital sits on two sides, and the simulator respects that.

Base leg

Simulated on-chain balances in the Base asset and USDC. Swaps consume and produce inventory on this side only.

Exchange leg

Simulated venue balances. A sell needs the asset already on the exchange — capital cannot teleport between legs mid-trade.

Rebalancing burden

Repeated one-directional trades drain one side. The simulator tracks the drift and the transfer cost a real operator would eventually pay.

Routes and sizes

The sweep space.

Directions evaluated

  • DEX → CEX

    Buy on-chain, sell into a centralized order book.

  • CEX → DEX

    Buy on a centralized order book, sell into Base liquidity.

  • DEX → DEX

    Two Base pools against each other, settled in one network.

Aerodrome → CoinbaseCoinbase → AerodromeUniswap v3 → KrakenKraken → Uniswap v3Aerodrome → Uniswap v3

Trade sizes swept per route

Profitability is size-dependent: the same spread can survive at one notional and be destroyed by impact at another. Every route is evaluated at each size independently.

$100
$250
$500
$1k
$2.5k
$5k
$10k

Configured notional ladder · USD-equivalent

Venues

Read-only connectivity by source.

  • AerodromeBase DEXOn-chain quoting through pool getAmountOut across vAMM and sAMM curves.Live
  • Uniswap v3Base DEXQuoter-based executable output, trialled across every deployed fee tier.Live
  • Coinbase AdvancedCEXDepth-walked order books. ETH is quoted against USD, so the USD/USDC basis is treated as an unmodelled term.Live
  • KrakenCEXDepth-walked order books on USDC-quoted markets where available.Live
  • OKXCEXDepth-walked order books, used as a breadth source for expanded markets.Live
  • BinanceCEXRegion-restricted from our infrastructure (HTTP 403/451). The adapter self-parks and other venues carry the leg.Restricted

System surfaces

What is connected, and what deliberately is not.

  • Base network access

    Multi-endpoint RPC pool with failover and rate-limit backoff.

  • DEX market data

    On-chain executable quoting from Base pools.

  • CEX market data

    Depth-walked order books from connected venues.

  • Scanner

    Detect, confirm, simulate and persist loop.

  • Paper engine

    Two-leg simulation against a reconciled inventory ledger.

  • Live execution adapter

    Not connected. Real trading is disabled.

Terminal

The operator view.

Operator terminal · redacted previewSimulation
MarketRouteSizeEngine decision
ETH/USDCAerodrome → Coinbase$1,000Rejected · fees exceed spread
ETH/USDCCoinbase → Uniswap v3$2,500Rejected · depth at size
cbBTC/USDCAerodrome → Kraken$500Preflight · identity verified
USDT/USDCUniswap v3 → OKX$10,000Rejected · below safety buffer

Representative screen · values illustrative · no results published