The Agent Skills Directory

[SAFE]: The skill instructions and patterns exclusively focus on legitimate Web3 frontend development using standard industry libraries (Wagmi, Viem, React). No malicious intent or suspicious behavior was identified.\n- [DATA_EXPOSURE_&_EXFILTRATION]: No evidence of hardcoded credentials, sensitive file access, or unauthorized network communication. The configuration examples use environment variables for public project identifiers, which is standard practice.\n- [INDIRECT_PROMPT_INJECTION]: While the skill involves processing external data (Ethereum addresses), it implements strong defensive measures by mandating the use of viem's getAddress() function for validation and checksum correction before any interaction occurs.\n- [REMOTE_CODE_EXECUTION]: No remote script downloads or dynamic code execution patterns were found. All dependencies are standard, versioned packages from public registries.\n- [OBFUSCATION]: The skill uses clear, readable markdown and code snippets. No hidden characters, encoded payloads, or homoglyph substitutions were detected.

wagmi-v3