eval-harness Audit report.

Warnings

• Dependency or permission surface needs review
• Permission surface may require sandboxing
• No explicit security guidance for executing eval commands, though the skill itself is documentation-only and does not introduce risky behavior.
• The skill is tailored to OpenAI Codex, which may limit broader applicability, but it is still a reusable methodology.
• Quality score needs review
• Permission surface needs review: secrets or environment access, shell or command execution
• GitHub adoption: 89 GitHub stars
• Stars/forks activity: 89 stars, 2 forks; issue activity unavailable in current metadata
• Dependency/runtime risk: command execution surface, credential or environment access
• Permission surface: secrets or environment access, shell or command execution

Method

This report combines public metadata, AI review output, repository freshness, install readiness, OpenAgentSkill events, quality scoring, trust checks, and the agent safety gate. It is not a full source-code security review.