πŸ₯ͺ Sandwich Attack Simulator

Constant-product AMM price-impact & MEV extraction, run live in your browser
Grounded in "No Place to Hide: An Analysis on Protected Order Flow Sandwich Attacks" (Heimbach, Solmaz, Γ–z, Torres β€” arXiv:2609.28115). The paper runs a three-year measurement study across six chains and finds that trades submitted through "protected" channels β€” private RPCs and native front-running protections β€” still get sandwiched at massive scale: 28.0 million attacks on Solana alone, plus tens of thousands more on Ethereum, Tron and Base, via exposures like order-flow-auction leaks, reorged blocks, an RPC bug on Base, and validator/application-level leaks on Solana. Their conclusion: these protections give substantially weaker guarantees than users expect. Below, set a trade and watch a bot front-run it, let it execute at a worse price, then back-run for profit β€” computed live from real xΒ·y=k AMM math, not a canned number. Flip "private RPC" on and notice it only dents the attack rate, exactly as the paper found.
28.0M
Solana sandwiches
30,607
Ethereum sandwiches
38,567
Tron sandwiches
1,889
Base sandwiches
2,875
ETH reorg victims

Your Trade

Private RPC protection OFF β€” full mempool exposure
Pool price Attacker front-run / back-run Your actual execution Price you expected (no attack)

Live Price Impact

mempool β€” bot spotting your pending trade
Statusβ€”
Price you expectedβ€”
Price you actually gotβ€”
Your slippage lossβ€”
Attacker front-run sizeβ€”
Attacker profit this tradeβ€”

Running Tally β€” Simulated Session

0
trades submitted
0
successful sandwiches
$0
total value extracted from you