human-relations

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a domain-specific knowledge base for academic journal submission criteria and does not contain any executable code or suspicious instructions.
  • [DATA_EXFILTRATION]: No network operations or patterns associated with data harvesting were found. The skill directs the agent to search official journal guidelines using its standard tools.
  • [PROMPT_INJECTION]: The instructions are focused on providing a template for paper analysis and do not attempt to override the AI's system prompt or security constraints.
  • [COMMAND_EXECUTION]: There are no shell commands, script blocks, or dynamic context injections detected in the provided file.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 03:41 AM
Security Audit — agent-trust-hub — human-relations