security-reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions for invoking a subagent named 'security-reviewer'. The implementation uses a standard heredoc pattern for passing context to the subagent, which prevents unintended shell expansion of the provided text.
- [COMMAND_EXECUTION]: The skill utilizes the
codex-subagentcommand to delegate tasks to a specialized agent card. This is a legitimate pattern for multi-agent coordination within the platform environment. - Evidence:
codex-subagent security-reviewer <<'EOF'in SKILL.md.
Audit Metadata