crossframe
Pass
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a 'no-code' instructional set. It contains 172 files, primarily markdown documents, which provide structured mental models and reasoning steps for the AI agent.
- [SAFE]: No network-enabled tools or commands (such as curl, wget, or fetch) are utilized or instructed within the provided files. References to external sources like GitHub are used purely for documentation and license purposes.
- [SAFE]: No sensitive file paths or credential patterns (e.g., .ssh, .aws, API keys) are accessed or targeted for exfiltration.
- [SAFE]: The skill includes extensive internal 'Guardrails' and 'Anti-Misuse Rules' that explicitly instruct the agent to avoid personality judgment, respect the responsibility chain, and protect low-power subjects, which enhances the overall safety and reliability of the agent's output.
- [SAFE]: Analysis of the markdown content for obfuscation, including Base64, homoglyphs, or zero-width characters, returned no suspicious results.
Audit Metadata