mw-replication

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill contains instructional text and structural templates for academic reproducibility in social sciences.- [COMMAND_EXECUTION]: The document references commands such as 'ssc install', 'pip install', and 'renv::snapshot()'. These are presented as best-practice examples for researchers to include in their own project documentation to ensure environment stability; they are not executed by the agent's runtime.- [DATA_EXFILTRATION]: No network requests, sensitive file access, or exfiltration patterns were found. The file path references are standard for research data organization (e.g., 'data/raw/', 'data/final/').
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 09:49 AM
Security Audit — agent-trust-hub — mw-replication