subagent-driven-review
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill utilizes platform-provided tools such as
rp_search,rp_abstracts,rp_cite, andrp_fulltextto interface with academic databases. These tools access well-known and reputable services including Scopus and OpenAlex. - [SAFE]: The instructions explicitly discourage insecure practices, advising against shell-level commands or manual network requests in favor of built-in tooling.
- [SAFE]: No obfuscation, hardcoded credentials, or unauthorized persistence mechanisms were found in the skill instructions or templates.
- [SAFE]: The multi-agent workflow is designed to manage context limits and optimize reasoning, representing standard and safe operational patterns for research-oriented agents.
Audit Metadata