The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an inherent attack surface because it is designed to ingest and process external smart contract code, which could potentially contain malicious instructions.
Ingestion points: Smart contract source files (e.g., src/Token.sol) are imported and compiled by the testing frameworks.
Boundary markers: The provided templates do not include specific delimiters or warnings to ignore instructions embedded in the contract code.
Capability inventory: The skill utilizes Hardhat and Foundry, which involve subprocess execution (npx hardhat test, forge test), file system writes (artifacts, coverage reports), and network operations (mainnet forking via RPC).
Sanitization: No source-level sanitization is performed on the contracts being tested.
[COMMAND_EXECUTION]: The skill provides templates for executing standard blockchain development and testing commands.
Evidence: Examples include npx hardhat test, npx hardhat coverage, and npx hardhat verify. These are standard tools in the Web3 ecosystem.
[CREDENTIALS_SAFE]: The skill follows security best practices for credential management.
Evidence: Hardhat configuration (hardhat.config.js) correctly utilizes environment variables (e.g., process.env.PRIVATE_KEY, process.env.ETHERSCAN_API_KEY) rather than hardcoding secrets.

web3-testing