viem
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- [NO_CODE] (SAFE): The skill consists entirely of Markdown documentation files. No executable scripts (.sh, .py, .js), binary files, or active configuration files are included.\n- [CREDENTIALS_UNSAFE] (SAFE): Code examples use standard placeholders for sensitive data (e.g., '0x...', 'legal winner...') and provide explicit security guidance against hardcoding private keys in source code.\n- [PROMPT_INJECTION] (SAFE): No instructions designed to override agent behavior, bypass safety filters, or reveal system prompts were found in the metadata or documentation body.\n- [DATA_EXFILTRATION] (SAFE): No network-capable tools or exfiltration patterns were detected. External links are limited to legitimate project documentation (viem.sh) and official GitHub repositories.\n- [INDIRECT_PROMPT_INJECTION] (SAFE): While the skill describes how to interact with external blockchain data (an inherent risk surface for Ethereum tools), the skill itself is purely informational and does not provide an exploitable tool interface or automated data processing script.
Audit Metadata