deep-research
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, command execution, or data exfiltration detected. The skill consists entirely of natural language instructions for organizing research tasks.
- [SAFE]: The search strategy recommends using well-known and trusted domains such as Arxiv and GitHub for sourcing information.
- [SAFE]: The skill does not include any executable code, external scripts, or package dependencies.
- [SAFE]: While the skill ingests untrusted data from the web (scraping), which is a standard attack surface for research tools, it includes instructions for fact-checking and cross-referencing to verify the accuracy of findings.
Audit Metadata