mgsci-topic-selection

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill consists entirely of Markdown-based instructions and decision frameworks for research topic selection. No executable Python, JavaScript, or shell scripts are present in the provided file.
  • [SAFE]: No malicious patterns such as prompt injection, credential exfiltration, or obfuscation were detected. The instructions are aligned with the stated purpose of academic fit analysis.
  • [SAFE]: The reference to resources/official-source-map.md indicates a local file read operation for retrieving editorial rules. Since no network operations or command execution capabilities are present, this does not pose a security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 11:57 PM
Security Audit — agent-trust-hub — mgsci-topic-selection