The Agent Skills Directory

[SAFE]: The skill demonstrates safe credential management by utilizing dotenv and environment variables (process.env) for the Alchemy API key and the deployer's private key, ensuring no secrets are hardcoded in the repository.
[SAFE]: External dependencies installed via npm and npx are standard, well-known packages within the Web3 development community, such as Hardhat and the Alchemy SDK.
[INDIRECT_PROMPT_INJECTION]: The skill interacts with external blockchain data via Alchemy RPC, which constitutes an ingestion surface for untrusted data.
Ingestion points: Smart contract data in test/fork-test.ts and balance/transaction data in test/helpers/alchemy-helper.ts.
Boundary markers: None explicitly defined in instructions to delimit data from instructions.
Capability inventory: The skill uses Bash (npm/npx) for project initialization and testing, and Write for configuration files.
Sanitization: Uses standard library parsing from ethers and alchemy-sdk for data types, but no specific prompt sanitization is applied to contract-retrieved strings.

alchemy-local-dev-loop