The Agent Skills Directory

[SAFE]: The skill provides documentation for the viem library, focused on Ethereum interactions. It contains only technical references and code examples for standard blockchain operations.
[CREDENTIALS_UNSAFE]: While the library facilitates the use of private keys and mnemonics, the documentation correctly advises against hardcoding them in source code and recommends secure practices like using environment variables. Example mnemonics provided are standard test placeholders.
[INDIRECT_PROMPT_INJECTION]: The skill documents functions that ingest external data from blockchain RPC providers. Ingestion points: getEnsName, getLogs, getTransaction. Boundary markers: None explicitly mentioned for agent prompt construction. Capability inventory: Network RPC calls and blockchain state-changing transactions via writeContract. Sanitization: Recommends normalize for ENS names.

viem