How Deployment Works

The BattleChain deploy stack — how just recipes, battlechain-lib, BattleChainDeployer, and CreateX fit together, plus the Foundry flags you need

When you deploy to BattleChain you're using a layered stack. Each layer is a thin wrapper over the one below it — so the just deploy-protocol you run in the quickstart and the bcDeployCreate(...) you write in your own script are the same path, just entered at different heights.

The stack

just recipe            →  starter-kit convenience wrapper (forge script ... --skip-simulation)
  forge script         →  your Solidity deploy script
    battlechain-lib    →  BCDeploy / BCSafeHarbor helpers (bcDeployCreate, createAndAdoptAgreement, requestAttackMode)
      BattleChainDeployer →  deploys + auto-registers the contract with the AttackRegistry
        CreateX        →  the underlying CREATE/CREATE2/CREATE3 factory

Layer	What it is	When you touch it
`just` recipes	Convenience aliases in the starter kit that run `forge script` with the right flags	Quickstart / fastest path
`battlechain-lib`	The library you install (`BCDeploy`, `BCSafeHarbor`, `BCScript`) exposing `bcDeployCreate*`, `createAndAdoptAgreement`, `requestAttackMode`	Writing your own deploy + agreement scripts
`BattleChainDeployer`	On-chain deployer that wraps CreateX and auto-registers each deployment with the `AttackRegistry` (so contracts are attackable)	Reference / direct calls
CreateX	Standard CREATE/CREATE2/CREATE3 factory used under the hood	Rarely directly

The key reason deployments go through BattleChainDeployer rather than a raw CREATE is registration: it records the deployer of every contract so the AttackRegistry can resolve which agreement covers it. Deploy outside this path and your contract won't be linked to its agreement.

Which layer should I use?

Just trying it out? Use the starter kit's just recipes — see Deploy and Battle-Test Your Contract.
Deploying your own contracts? Install battlechain-lib and write a script with BCScript / bcDeployCreate* — see Deploying Contracts.
Need the raw addresses/ABIs? See the Contract API Reference.

battlechain-lib works on BattleChain and 190+ other EVM chains (it falls back to CreateX directly off-BattleChain). Only requestAttackMode is BattleChain-specific.

Foundry on BattleChain

These apply to every forge script / cast call against BattleChain. The starter kit's justfile and battlechain-lib's battlechain.just already include them.

Always pass `--skip-simulation`

Forge's local gas estimation isn't reliable on BattleChain, so simulation can fail before broadcast. Skip it:

forge script script/Deploy.s.sol --rpc-url https://testnet.battlechain.com --broadcast --skip-simulation

Add `-g 300` if a deploy still fails

Bumps the gas estimate to 3× to clear vague out-of-gas failures:

forge script script/Deploy.s.sol --rpc-url https://testnet.battlechain.com --broadcast --skip-simulation -g 300

`--legacy` is not needed

BattleChain accepts standard EIP-1559 transactions — none of the working deploy scripts use --legacy. If a tool or older guide tells you to add it, you can ignore that.

Watch the 24 KB contract size limit

BattleChain enforces the EVM's 24,576-byte limit. An oversized contract can't be gas-estimated, which surfaces as:

AnyTxType(2) transaction can't be built due to missing keys: ["gas_limit"]

Fix it by enabling the optimizer, splitting the contract, or compiling with --via-ir. Check sizes with forge build --sizes.

Deploy and Battle-Test Your Contract

Put the stack to work — deploy a vault and open it for attack