scientific-problem-selection
Pass
Audited by Gen Agent Trust Hub on Nov 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a purely instructional framework for an AI agent to act as a scientific research advisor. It consists of markdown instructions and reference materials based on established scientific literature.
- [SAFE]: No executable code, scripts, or system commands were found. The skill operates within the standard conversational capabilities of the agent.
- [SAFE]: References to external tools, such as PubMed searches, are for information retrieval purposes consistent with the skill's stated goal of scientific validation and are not misused for exfiltration or malicious intent.
- [SAFE]: No obfuscation, data exfiltration patterns, or persistence mechanisms were detected.
Audit Metadata