hrm-data-analysis

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and provides methodological framework for academic research. No prompt injections, data exfiltration patterns, or obfuscated content were detected.
  • [COMMAND_EXECUTION]: The skill mentions specific statistical analysis functions (e.g., romano_wolf, oster_delta, wild_cluster_bootstrap) intended for use via Model Context Protocol (MCP) handles. These are domain-specific research tools and do not represent arbitrary or dangerous shell execution.
  • [EXTERNAL_DOWNLOADS]: No remote scripts or binary downloads are present. References to 'execution bridges' point to local relative paths within the project structure (e.g., ../../../shared-resources/), which are part of the local skill environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 09:43 AM
Security Audit — agent-trust-hub — hrm-data-analysis