s4h-cognition-mental-models

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: No malicious patterns or security risks detected. The skill provides a structured framework for cognitive analysis through natural language instructions.- [NO_CODE]: The skill contains no executable scripts, binaries, or package dependencies. All logic is implemented via prompt instructions.- [SAFE]: The skill interacts with the user via the AskUserQuestion tool, which is a restricted interaction capability with no access to the file system, network, or sensitive environment variables.- [SAFE]: Analysis of the markdown content found no evidence of prompt injection, obfuscation, or data exfiltration attempts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 11:43 PM
Security Audit — agent-trust-hub — s4h-cognition-mental-models