human-ai
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a local Python script,
scripts/measure.py, used to calculate linguistic metrics such as Type-Token Ratio (TTR), burstiness, and Shannon entropy. This script relies exclusively on Python standard libraries and is executed locally to provide deterministic analysis. - [EXTERNAL_DOWNLOADS]: The skill documentation describes a discovery protocol for integrating the Jev (TypeSafe AI) model via OpenRouter or MCP tools. This is a documented extension for calibrated scoring and does not involve unauthorized remote code execution or data exfiltration outside of the intended model evaluation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted text input to perform stylistic rewriting. It incorporates diagnostic checks for common AI-slop patterns and maintains a focused scope on editorial transformation, which mitigates the risk of instructions within the processed text influencing the agent's behavior.
Audit Metadata