skill-08-security-audit
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a defensive tool designed for security auditing of smart contracts. It defines legitimate workflows for evaluating contract safety and does not contain any malicious code or instructions.
- [COMMAND_EXECUTION]: The skill references specific Model Context Protocol (MCP) servers (
blockchain-rpc-server,defi-protocol-server) for executing blockchain-related tasks. These tools are used appropriately for its stated purpose of security analysis. - [DATA_EXPOSURE_AND_EXFILTRATION]: No sensitive file access, credential harvesting, or unauthorized data transmission patterns were detected. The data processed involves public blockchain state and security databases.
- [PROMPT_INJECTION]: The instructions do not attempt to override the agent's system prompt, bypass safety filters, or use role-play to evade restrictions.
- [OBFUSCATION]: The skill content is clear and readable, with no encoded strings, hidden characters, or homoglyph substitutions identified.
Audit Metadata