ORA
Ask or search…
K
Links

User Guide

If you want your CLE to use ZK Automation, then you need to make two changes to the CLE and set Bounty when deploying:

1. Set dataDestinations in cle.yaml

In a CLE that uses ZK Automation, you need to first define the onchain smart contract that will be called when automation should be triggered.
For example:
specVersion: 0.0.2
apiVersion: 0.0.2
# ...
dataSources:
# ...
# dataDestinations field with valid contract address is required for ZK Automation.
dataDestinations:
- kind: ethereum
network: sepolia
address: "0xdeadbeef000000000000000000000000000000000"
Note that if you don't need to use ZK Automation, you don't need to set dataDestinations or just set the address to be something like burner address.

2. Return Payload (Calldata) in mapping.ts

In the previous step, we have defined the blockchain network triggered by automation and the smart contract.
The role of mapping.ts:
  • Determining whether automation needs to triggered: When handleBlocks does not return a value (eg. exit due to false condition in require), the automation does not need to be triggered.
  • Setting the payload (calldata) for the trigger: when handleBlocks returns value, the automation needs to be triggered, and the return value is used as the payload (calldata) for the call. If you are confused about these concepts, please check introduction page of CLE.
For example:
This CLE's ZK Automation will never be triggered.
export function handleBlocks(blocks: Block[]): Bytes {
// Always exit at this line.
require(false);
return Bytes.empty();
}
This CLE's ZK Automation will always be triggered, and function of run() will be called.
export function handleEvents(blocks: Block[]): Bytes {
// run() func signature - c0406226
// zkAuto call input data onchain - 0xc0406226
return Bytes.fromHexString("c0406226").padEnd(32);
}

3. Set Bounty when Publishing

Since each proof generation and onchain call (including verification and later calls to dataDestinations) consumes fund, we set up the Bounty mechanism.
Bounty mechanism means that the user of ZK Automation will deposit the bounty into our contract in advance, and set the amount of bounty that the ZK Oracle node can draw from each time ZK Automation is triggered at the time of deployment.
Currently in the testnet phase, you just need to set the bounty_reward_per_trigger when publishing CLE:
# Publish Command in CLE CLI

clepublish <ipfs_hash> <bounty_reward_per_trigger>
Check out this example of using ZK Automation from CLE scaffold.