What problem does it solve? AI agents trained on EVM patterns generate incorrect StarkNet code because StarkNet is not EVM-compatible: it uses Cairo instead of Solidity, felt252 instead of uint256, contract-based accounts instead of EOAs, and a two-step declare-then-deploy flow. This Skill provides correct, current patterns for Cairo contract development, testing, deployment, and cross-layer messaging. ## Core Features & Use Cases - Cairo Contract Patterns: Provides working examples of storage, interfaces, events, ownable contracts, OpenZeppelin components, and ERC-20 tokens using Cairo 2.x syntax. - Full Development Workflow: Covers Scarb project setup, snforge testing with cheatcodes, and deployment via starkli, sncast, or starknet.js, including account deployment. - L1-L2 Messaging: Documents bidirectional Ethereum-StarkNet messaging with Cairo send_message_to_l1_syscall, Solidity sendMessageToL2, and #[l1_handler] receivers. - Use Case: Ask the agent to write an ERC-20 token on StarkNet, and it produces a correct OpenZeppelin component-based Cairo contract, compiles it with Scarb, tests it with snforge, and deploys it to Sepolia with sncast. ## Quick Start Ask the agent to create a new Cairo counter contract on StarkNet, test it with snforge, and deploy it to the Sepolia testnet.