🦄 🤖 A replica implementation of an MEV honeypot bot I caught out in the wild.
First, prepare your .env.test file:
# A way of identifying the chain being targeted.
# This is useful for the state recovery mechanism, if an attack
# is terminated unsuccessfully.
CHAIN="mainnet-fork"
# The RPC we use to consult for blockchain state. In the test environment,
# this will be the address of your local `anvil` RPC, spun up with a call
# to `yarn fork`:
ETH_RPC_URL="http://127.0.0.1:8545"
# The Ethereum RPC for calls to `yarn fork` to fork from,
# i.e. Alchemy or Infura.
ETH_FORK_URL=""
# The mnemonic of the account which will be used to initialize the attack.
# In a long-term attack, this would be the `EXIT_MNEMONIC` of the attack
# which had just finished.
DEPLOYER_MNEMONIC=""
# The mnemonic of the account which will have funds transferred
# to, once the attack is finished.
# The idea here is we'd be able to spin up a simple high-level script
# which manages mnemonic generation and the transition between new
# attack pools.
EXIT_MNEMONIC=""
# Whether to generate fake interest in the token during tests.
# Useful for testing token rehypothecation. To appreciate how
# the attack works, you'll want this set to `true`.
SIMULATE_APPROVALS="true"
# An Etherscan key, used to programmatically verify.
ETHERSCAN_API_KEY=""Okay, in one window, run the following:
yarn forkThis spins up a local anvil fork of Ethereum Mainnet. In a separate window, run:
NODE_ENV=test yarn startTo load up the GUI and initialize the attack. You'll be asked to provide some high-level configuration properties like the token name, totalSupply() etc.
The attack CLI is command based, where you need to supply high-level commands to define specific operations. You type these using the green cursor once the attack has successfully initialized.
Tip
Piggyback will save your attack state to ensure that, if for whatever reason an attack is exited prematurely, it can be resumed.
This can cause some trouble with anvil forks, so when testing you might need to remove the auto-generated backup files which are populated in the file root.
# To swap ETH for Honeypot Tokens in the Pool,
# enter the following into the command line:
swapEth("0.1")\n
# To swap Honeypot tokens back, use:
swapHoneypotToken("0.1")\n
# To rug the initial liquidity from the pool, use:
rug()\n
# To manually steal Honeypot tokens back from any
# trading bots that were lured in, use (this is to
# avoid any bot from attempting to trade using the
# Honeypot token, to ensure the pool looks like
# buyers are truly trying to hold, whereas in reality
# they become programmatically insolvent.
refreshApprovals()\n
# To shut down the attack and migrate all remaining funds
# to the exit wallet, call:
finalize()\n
You can also run the following to sanity test the project:
yarn test
yarn piggyback