rw-research-referee

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a structured framework for scientific auditing of research designs, protocols, and statistics without requesting elevated permissions or external tool access.
  • [EXTERNAL_DOWNLOADS]: The skill references official documentation and guidelines from well-known scientific organizations including Cochrane (methods.cochrane.org), the EQUATOR Network (equator-network.org), and ICMJE (icmje.org). These references are used for methodology verification and are sourced from trusted academic domains.
  • [COMMAND_EXECUTION]: A bundled Python script (scripts/self_check.py) is included to perform integrity checks on the skill's local files. The script logic is restricted to verifying file existence and checking for the absence of hardcoded developer-specific local paths, posing no security risk to the user environment.
  • [DATA_EXFILTRATION]: The instructions explicitly mandate the separation of user-provided materials from inferences and external sources, and the skill is designed to run independently of private local workspaces or personal knowledge bases.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 10:37 PM
Security Audit — agent-trust-hub — rw-research-referee